Cross Reference: /src/lib/libedit

History log of /src/lib/libedit
Revision	Date	Author	Comments
1.1	30-Aug-2021	christos	Add a changes file
1.70	03-Aug-2023	rin	Revert CC_WNO_USE_AFTER_FREE from Makefile's (thanks uwe@)
1.69	03-Aug-2023	rin	Sprinkle CC_WNO_USE_AFTER_FREE for GCC 12 All of them are blamed for idiom equivalent to: newbuf = realloc(buf, size); p = newbuf + (p - buf);
1.68	20-Jun-2023	wiz	install pkg-config file for libedit version number matches portable libedit --cflags output matches portable libedit, since users probably want the readline interface
1.67	03-Jun-2023	lukem	bsd.own.mk: rename GCC_NO_* to CC_WNO_* Rename compiler-warning-disable variables from GCC_NO_warning to CC_WNO_warning where warning is the full warning name as used by the compiler. GCC_NO_IMPLICIT_FALLTHRU is CC_WNO_IMPLICIT_FALLTHROUGH Using the convention CC_compilerflag, where compilerflag is based on the full compiler flag name.
1.66	13-Oct-2019	mrg	introduce some common variables for use in GCC warning disables: GCC_NO_FORMAT_TRUNCATION -Wno-format-truncation (GCC 7/8) GCC_NO_STRINGOP_TRUNCATION -Wno-stringop-truncation (GCC 8) GCC_NO_STRINGOP_OVERFLOW -Wno-stringop-overflow (GCC 8) GCC_NO_CAST_FUNCTION_TYPE -Wno-cast-function-type (GCC 8) use these to turn off warnings for most GCC-8 complaints. many of these are false positives, most of the real bugs are already commited, or are yet to come. we plan to introduce versions of (some?) of these that use the "-Wno-error=" form, which still displays the warnings but does not make it an error, and all of the above will be re-considered as either being "fix me" (warning still displayed) or "warning is wrong."
1.65	30-Jun-2017	kre	branches: 1.65.6; Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale).
1.64	27-Jun-2017	christos	add literal escape sequence support, patterned after the tcsh ones.
1.63	24-May-2016	christos	branches: 1.63.8; remove debug read (Ingo Schwarze)
1.62	09-May-2016	christos	Instead of compiling all the source files together in one big file, use protected visibility to achieve the same effect.
1.61	02-May-2016	wiz	Add missing backslash that broke build.
1.60	02-May-2016	christos	Add more MLINKS, sort
1.59	28-Apr-2016	christos	new man page from Ingo Schwarze.
1.58	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.57	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.56	02-Mar-2016	christos	PR/50880: David Binderman: Remove redundant code. While here, fix all debugging formats.
1.55	24-Feb-2016	christos	Get split el_getc and el_wgetc completely and call el_wgetc internally. Change some character constants to they wide versions. (Ingo Schwarze)
1.54	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.53	29-Jan-2015	joerg	Disable -Wcast-qual for clang for now.
1.52	14-Jun-2014	mrg	branches: 1.52.2; remove remaining makefile support for GCC < 45 that i found.
1.51	10-Aug-2012	joerg	branches: 1.51.2; 1.51.8; Don't depend on HAVE_GCC being always defined.
1.50	21-Mar-2012	matt	These directories default to WARNS?=5
1.49	16-Aug-2011	christos	branches: 1.49.2; re-enable -Wconversion
1.48	02-Aug-2011	joerg	Only use -Wconversion if GCC is actually the active compiler.
1.47	30-Jul-2011	tron	Don't use "-Wconversion" with GCC 4.5 which will complain about all the expressions where signed variables are converted to unsigned in an expression e.g. "size_t foo = sizeof(something) * int_var;".
1.46	29-Jul-2011	christos	pass -Wconversion
1.45	29-Jul-2011	christos	add -Wunused-parameter Is that the right way? Perhaps WARNS=5?
1.44	28-Jul-2011	christos	Rename key to keymacro to avoid conflicts with term.h. The renaming of term to terminal was again to avoid conflicts with term.h. term.h is a moving namespace violation.
1.43	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.42	21-Jun-2011	mrg	add some XXX'd -Wno-foo if HAVE_GCC >= 45. XXX: someone should look at these.
1.41	03-Feb-2010	roy	branches: 1.41.4; Userland now builds and uses terminfo instead of termcap. OK: core@, jdc@
1.40	31-Dec-2009	christos	- Document and enable wide character support. - Fix read function compatibility.
1.39	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.38	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.37	18-Jan-2009	lukem	WARNS=4
1.36	28-May-2007	tls	Add new Makefile knob, USE_FORT, which extends USE_SSP by turning on the FORTIFY_SOURCE feature of libssp, thus checking the size of arguments to various string and memory copy and set functions (as well as a few system calls and other miscellany) where known at function entry. RedHat has evidently built all "core system packages" with this option for some time. This option should be used at the top of Makefiles (or Makefile.inc where this is used for subdirectories) but after any setting of LIB. This is only useful for userland code, and cannot be used in libc or in any code which includes the libc internals, because it overrides certain libc functions with macros. Some effort has been made to make USE_FORT=yes work correctly for a full-system build by having the bsd.sys.mk logic disable the feature where it should not be used (libc, libssp iteself, the kernel) but no attempt has been made to build the entire system with USE_FORT and doing so will doubtless expose numerous bugs and misfeatures. Adjust the system build so that all programs and libraries that are setuid, directly handle network data (including serial comm data), perform authentication, or appear likely to have (or have a history of having) data-driven bugs (e.g. file(1)) are built with USE_FORT=yes by default, with the exception of libc, which cannot use USE_FORT and thus uses only USE_SSP by default. Tested on i386 with no ill results; USE_FORT=no per-directory or in a system build will disable if desired.
1.35	31-Aug-2006	rpaulo	Rename TEST/test.c to avoid problems when doing a cleandir on case-insensitive file systems. ok'ed christos.
1.34	28-May-2005	lukem	clean up build of "test"
1.33	28-May-2005	lukem	MAKEVERBOSE support
1.32	18-May-2005	christos	Libedit depends on libterm. From Patrick Welche
1.31	07-May-2005	dsl	Make everything that uses makelist depend on Makefile - that way the created files pick up new entries.
1.30	07-May-2005	dsl	Separate out the filename completion functions from the readline() code. Pass in loads of parameters instead of relying on shed-loads of global variables to modify the behaviour. The filename completion code can now be enabled by code that uses el_gets(). (eg /bin/sh)
1.29	05-Dec-2003	lukem	Tokenization function enhancements: * Make tok_init(), tok_end(), tok_reset(), tok_line() and tok_str() publically available in <histedit.h> * Documented the public functions in editline(3) * Renamed tok_line() -> tok_str() * Added new tok_line() which takes a "const LineInfo " instead of "const char " (the former has "cursor" information), and optionally return the argv index ("int cursorc") and offset within that index ("int cursorv"). This means that completion routines can use the tokenization code to crack the line and easily find which word the cursor is at. (mmm, context sensitive completion :) * Fixed TEST/test.c when using "continuation" lines (unmatched quote or \ at EOL), and added some more DEBUG messages including highlighting where the cursor is (with a `_').
1.28	01-Aug-2003	lukem	Rework how dependency generation is performed: * DPSRCS contains extra dependencies, but is _NOT_ added to CLEANFILES. This is a change of behaviour. If a Makefile wants the clean semantics it must specifically append to CLEANFILES. Resolves PR toolchain/5204. * To recap: .d (depend) files are generated for all files in SRCS and DPSRCS that have a suffix of: .c .m .s .S .C .cc .cpp .cxx * If YHEADER is set, automatically add the .y->.h to DPSRCS & CLEANFILES * Ensure that ${OBJS} ${POBJS} ${LOBJS} ${SOBJS} .d depend upon ${DPSRCS} Deprecate the (short lived) DEPENDSRCS Update the various Makefiles to these new semantics; generally either adding to CLEANFILES (because DPSRCS doesn't do that anymore), or replacing specific .o dependencies with DPSRCS entries. Tested with "make -j 8 distribution" and "make distribution".
1.27	31-Jul-2003	lukem	tweak this unconventional (some might say "baroque") Makefile to work with the new <bsd.dep.mk> *.d semantics. fixes problems highlighted by Martin Husemann <martin@>
1.26	08-May-2003	christos	add a missing dependency (John Gordon)
1.25	08-May-2003	christos	Use ${HOST_SH}
1.24	27-Oct-2002	christos	vi mode and memory fixes from david laight.
1.23	19-Aug-2002	lukem	Explicitly move setting of NOxxx and USE_SHLIBDIR to the top of the Makefile (before including <bsd.own.mk>)
1.22	18-Mar-2002	christos	Add a couple linted comment and enable WARNS=3
1.21	28-Dec-2001	lukem	* Add user-controlled mk.conf variables - SHLIBDIR Location to install shared libraries if ${USE_SHLIBDIR} is "yes". Defaults to "/usr/lib". - USE_SHLIBDIR If "yes", install shared libraries in ${SHLIBDIR} instead of ${LIBDIR}. Defaults to "no". Sets ${_LIBSODIR} to the appropriate value. This may be set by individual Makefiles as well. - SHLINKDIR Location of shared linker. Defaults to "/usr/libexec". If != "/usr/libexec", change the dynamic-linker encoded in shared programs * Set USE_SHLIBDIR for libraries used by /bin and /sbin: libc libcrypt libcrypto libedit libipsec libkvm libm libmi387 libtermcap libutil libz * If ${_LIBSODIR} != ${LIBDIR}, add symlinks from ${LIBDIR}/${LIB}.so* to ${_LIBSODIR}/${LIB}.so* for compatibility. * Always install /sbin/init statically (for now) The net effect of these changes depends on how the variables are set: 1.) If nothing is set or changed, there is no change from the current behaviour: - Static /bin, /sbin, and bits of /usr/* - Dynamic rest - Shared linker is /usr/libexec/ldso 2.) If the following make variables are set: LDSTATIC= SHLINKDIR=/lib SHLIBDIR=/lib Then the behaviour becomes: - Dynamic tools - .so libraries used by /bin and /sbin are installed to /lib, with symlinks from /usr/lib/libso to -> /lib/libso where appropriate - Shared linker is /lib/ldso 3.) As per 2.), but add the following variable: USE_SHLIBDIR=yes This forces all .so's to be instaleld in /lib (with compat symlinks), not just those tagged by their Makefiles to be. Again, compat symlinks are installed
1.20	05-Jan-2001	jdolecek	Standard location of readline headers is /usr/include/readline/, so install them there. readline.h of libedit had to move to subdirectory 'readline', due to the way BSD makefiles work; this is better than potentially fragile Makefile hacks
1.19	15-Aug-2000	mrg	use .tmp temporaries for generated files, to avoid having failed generated output being used.
1.18	06-Jul-1999	christos	Use LIBEDITDIR instead of CURDIR so we can use that Makefile to compile libedit from another directory.
1.17	02-Jul-1999	simonb	More trailing white space.
1.16	04-Mar-1999	itohy	Add minimal dependency to make "make depend" optional after cleandir.
1.15	29-Jul-1998	lukem	* add more checks for NULL pointers in passed arguments * implement el_get(EditLine , int op, void result), which does the inverse of el_set() * add EL_EDITMODE operation to el_set and el_get; if non zero editing is enabled (the default). * add "edit on \| off" editrc command, which modifies EL_EDITMODE. users can now add '*:edit off' in ~/.editrc as an advisory to disable editing. NOTE: at this time EL_EDITMODE is just an indication of the state of the 'edit' command. It's up to the application to check this after el_source() or el_parse() to determine if editing is still required.
1.14	13-Nov-1997	thorpej	Build readline.o into debugging versions of libedit, too.
1.13	26-Oct-1997	christos	Make a link for history.h -> readline.h
1.12	23-Oct-1997	christos	PR/4301: Jaromir Dolecek. Add gnu-readline wrapper for editline.
1.11	23-Oct-1997	lukem	use CPPFLAGS instead of CFLAGS, fix building of test
1.10	09-Oct-1997	lukem	branches: 1.10.2; - define WARNS?=1 in the top-level Makefile.inc, and don't define anywhere else. - for now, override WARNS=0 in librpcsvc and libwrap, until they're cleaned up - rcsid police lib is now clean (except for librpcsvc and libwrap) on the i386, and this should motivate the other ports to fix any other minor problems that their compilers pick up that the i386 version doesn't.
1.9	06-Jul-1997	christos	Fix compiler warnings.
1.8	09-May-1997	mycroft	Eliminate bogus redefinitions of standard targets.
1.7	24-Mar-1997	christos	Makefile cleanups: use INCS variable to install includes and FILES to install objects.
1.6	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.5	09-Jan-1997	lukem	* add a man page for the editline routines * add a man page describing editrc * fix bugs in el_parse(): * didn't execute command when program name matched (test reversed) * was checking against empty string instead of program name * after checks, command to run also pointed to empty string [christos - the author of libedit - ok-ed the man pages in general (which I wrote from scratch by RTFS) as well as the bugfix]
1.4	18-Oct-1996	thorpej	Use ${INSTALL}.
1.3	01-Jun-1996	jtk	merge bugfix from 1.2 branch: use includes target for include files
1.2	06-May-1994	cgd	branches: 1.2.8; local
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.2.8.1	01-Jun-1996	jtk	use includes target to install include files
1.10.2.1	09-Nov-1997	lukem	sync trunk's CPPFLAGS fix (approved by thorpej)
1.41.4.1	23-Jun-2011	cherry	Catchup with rmind-uvmplock merge.
1.49.2.2	30-Oct-2012	yamt	sync with head
1.49.2.1	17-Apr-2012	yamt	sync with head
1.51.8.1	10-Aug-2014	tls	Rebase.
1.51.2.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.52.2.1	13-May-2015	martin	Sync lib/libedit with head, requested by christos in #753: lib/libedit/Makefile 1.53 lib/libedit/chartype.h 1.13 lib/libedit/editline.3 1.83-1.84 lib/libedit/editrc.5 1.28-1.29 lib/libedit/eln.c 1.18 lib/libedit/filecomplete.c 1.33-1.34 lib/libedit/readline.c 1.112-1.115 Man page improvements, fix overlapping strcpy, improve readline compatibility, clang build fix.
1.63.8.1	23-Jul-2017	snj	Pull up following revision(s) (requested by kre in ticket #102): lib/libedit/Makefile: 1.64-1.65 lib/libedit/editline.3: 1.94-1.96 lib/libedit/editrc.5: 1.33 lib/libedit/el.c: 1.93-1.94 lib/libedit/el.h: 1.42 lib/libedit/literal.c: 1.1-1.3 lib/libedit/literal.h: 1.1-1.2 lib/libedit/prompt.c: 1.27 lib/libedit/read.c: 1.103 lib/libedit/refresh.c: 1.52-1.54 lib/libedit/refresh.h: 1.11 lib/libedit/terminal.c: 1.33 Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming. -- Include EDITRC in doc. -- mention the limitation of the literal sequence delimiter. -- - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display. -- - add literal sequence handling. -- remove unused variable -- add literal escape sequence support, patterned after the tcsh ones. -- fix comment -- Fix an obvious, but almost invisible typo (avoid some core dumps). -- Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale). -- Remove workaround for ancient HTML generation code.
1.65.6.1	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.1	02-Apr-2013	agc	branches: 1.1.2; file Version.map was initially added on branch agc-symver.
1.1.2.1	02-Apr-2013	agc	add symbol versioning info for: libcrypt libcurses libedit libmenu libossaudio libposix libskey
1.64	29-Jun-2024	christos	Prevent testing out of bounds memory. From Robert Morris https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=279762
1.63	30-Oct-2022	christos	improvements in malloc/free handling.
1.62	08-Feb-2022	rillig	libedit: fix typos, apply KNF to newly imported code (PR/56693) No binary change.
1.61	08-Feb-2022	christos	PR/56693: Walter Lozano: Add support for rl_delete_text and rl_set_key
1.60	11-Jan-2022	christos	PR/56618: Walter Lozano: Improve libedit compatibility with readline by implementing: rl_copy_text, rl_erase_empty_line, rl_message, rl_on_new_line, rl_replace_line, rl_restore_prompt, rl_save_prompt
1.59	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.58	23-Jul-2019	christos	PR/54400: S�ren Tempel: out-of-bounds read in libedit c_delbefore
1.57	11-Oct-2017	abhinav	branches: 1.57.4; Fix typo in comment
1.56	22-May-2016	christos	Stop the read module from poking the el_chared.c_macro data structure currently belonging to the chared module. The read module does so from three of its functions, while no other module uses the macro data, not even the chared module itself. That's quite logical because macros are a feature of input handling, all of which is done by the read module, and none by the chared module. So move the data into the read modules's own opaque data structure, struct el_read_t. That simplifies internal interfaces in several respects: The semi-public chared.h has one fewer struct, one fewer #define, and one fewer member in struct el_chared_t; all three move to one single C file, read.c, and are now module-local. And the internal interface function ch_reset() needs one fewer argument, making the code of many functions in various modules more readable. The price is one additional internal interface function, read_end(), 10 lines long including comments, called publicly from exactly one place: el_end() in el.c. That's hardly an increase in complexity since most other modules already have their *_end() function, read.c was the odd one out not having one. From Ingo Schwarze
1.55	09-May-2016	christos	s/protected/libedit_private/g
1.54	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.53	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.52	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.51	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.50	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.49	24-Feb-2016	christos	A very simple, non-intrusive patch to fix a segfault (and a functional error) in c_gets(), file chared.c. Run any program using libedit in the default way. At the el_[w]gets() prompt, invoke ed-command (for example, in emacs mode, press the escape key, then type the letter 'x'). You should see a ": " prompt. Type the letter 'x' again. Now press the backspace key a few times, looking at the screen after each key press: - The 1st BS deletes the 'x'. - The 2nd BS deletes the blank after the prompt. - The 3rd BS deletes the colon of the prompt. - The 4th BS moves the cursor up one line. - The 5th BS gives me "Segmentation fault (core dumped)". Depending on your platform, it might take a few more or a few less backspaces for the buffer underrun to trigger the segfault, but you should be able to hit it sooner or later no matter what. Run the same program again, connect again and invoke ed-command again. Now type: 'b', backspace, 'i', backspace, 'n', backspace, 'd', enter. The "bind" command gets executed, even though you deleted what you typed before hitting enter. From Ingo Schwatze.
1.48	24-Feb-2016	christos	Get split el_getc and el_wgetc completely and call el_wgetc internally. Change some character constants to they wide versions. (Ingo Schwarze)
1.47	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.46	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.45	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.44	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.43	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.42	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.41	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.40	18-Jun-2014	christos	Don't depend on weak aliases to define the vi "alias" expansion function, provide an API instead to set it.
1.39	12-Jul-2013	christos	branches: 1.39.4; cast to avoid warning.
1.38	12-Jul-2013	christos	Add a function to move the cursor.
1.37	18-Jul-2012	christos	branches: 1.37.2; From Kamil Dudka: fix crash of el_insertstr() on incomplete multi-byte
1.36	23-Oct-2011	christos	branches: 1.36.2; Fixed misplaced parenthesis (Nirbhay Choubey)
1.35	16-Aug-2011	christos	re-enable -Wconversion
1.34	29-Jul-2011	christos	pass -Wconversion
1.33	29-Jul-2011	christos	kill ifdef notdef
1.32	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.31	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.30	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.29	28-Aug-2010	christos	setup a callback to be invoked on resize buffers so that readline can reset rl_line_buffer which unfortunately some applications use it directly.
1.28	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.27	15-Feb-2009	christos	pass lint on _LP64.
1.26	06-Feb-2009	sketch	branches: 1.26.2; de-__P()
1.25	08-Aug-2005	christos	The previous commit removed too much and forgot to reset the history event number. From Kouichirou Hiratsuka, many thanks!
1.24	01-Aug-2005	christos	Don't reset the macro strings each time we enter el_gets(), otherwise el_push() is unusable programmatically.
1.23	01-Jun-2005	lukem	Don't use non-standard uint or u_int.
1.22	13-Aug-2004	mycroft	Delete-previous-char and delete-next-char without an argument are not supposed to modify the yank buffer in Emacs. Make it so.
1.21	02-Nov-2003	christos	Always use el->el_buffer, because newbuffer could have moved. From Gerry Swislow gerry at certif dot com
1.20	18-Oct-2003	christos	change allocation policy in el_push to allocate the string itself. fix issues with strdup.
1.19	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.18	20-Nov-2002	christos	Fix problem with previous patches that broke vi history. - c_gets() was usually returning a length, but sometimes one of the CC_xxx values (which are small +ve integers)! - fixed c_gets() by putting a ' ' under the cursor. From David Laight.
1.17	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.16	27-Oct-2002	christos	vi mode and memory fixes from david laight.
1.15	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.14	17-May-2001	christos	PR/12963:Jason Waterman: Fix signed cast problems.
1.13	13-Apr-2001	lukem	knf ch_enlargebufs(), to be consistent with the rest of this file...
1.12	10-Jan-2001	jdolecek	Enlarge editline buffers as needed to support arbitrary length lines. This also addresses lib/9712 by Phil Nelson.
1.11	04-Jan-2001	christos	consistently check for allocation failures and return -1, if we could not get more memory.
1.10	11-Nov-2000	christos	- add support for home and end keys. - improve debugging support
1.9	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.8	28-Feb-2000	chopps	el_insertstr takes a "const char " not "char " now as it doesn't modify the argument.
1.7	02-Jul-1999	simonb	More trailing white space.
1.6	05-Feb-1999	christos	delint.
1.5	12-Dec-1998	christos	delint
1.4	03-Feb-1998	perry	remove obsolete register declarations
1.3	06-Jul-1997	christos	Fix compiler warnings.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.26.2.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.36.2.2	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.36.2.1	30-Oct-2012	yamt	sync with head
1.37.2.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.39.4.1	10-Aug-2014	tls	Rebase.
1.57.4.1	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.30	22-May-2016	christos	Stop the read module from poking the el_chared.c_macro data structure currently belonging to the chared module. The read module does so from three of its functions, while no other module uses the macro data, not even the chared module itself. That's quite logical because macros are a feature of input handling, all of which is done by the read module, and none by the chared module. So move the data into the read modules's own opaque data structure, struct el_read_t. That simplifies internal interfaces in several respects: The semi-public chared.h has one fewer struct, one fewer #define, and one fewer member in struct el_chared_t; all three move to one single C file, read.c, and are now module-local. And the internal interface function ch_reset() needs one fewer argument, making the code of many functions in various modules more readable. The price is one additional internal interface function, read_end(), 10 lines long including comments, called publicly from exactly one place: el_end() in el.c. That's hardly an increase in complexity since most other modules already have their *_end() function, read.c was the odd one out not having one. From Ingo Schwarze
1.29	09-May-2016	christos	s/protected/libedit_private/g
1.28	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.27	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.26	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.25	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.24	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.23	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.22	18-Jun-2014	christos	Don't depend on weak aliases to define the vi "alias" expansion function, provide an API instead to set it.
1.21	28-Aug-2010	christos	branches: 1.21.12; 1.21.22; setup a callback to be invoked on resize buffers so that readline can reset rl_line_buffer which unfortunately some applications use it directly.
1.20	15-Apr-2010	christos	From Jess Thrysoee - Fix wint_t to Int confusion
1.19	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.18	15-Feb-2009	christos	pass lint on _LP64.
1.17	06-Mar-2006	christos	branches: 1.17.28; Print the actual eofc, instead of ^D\b\b. Change internal character decoding to prevent buffer oveflows.
1.16	08-Aug-2005	christos	Spelling mistakes and comment errors (from FreeBSD via Stefan Farfeleder; many thanks)
1.15	01-Aug-2005	christos	Don't reset the macro strings each time we enter el_gets(), otherwise el_push() is unusable programmatically.
1.14	13-Aug-2004	mycroft	Delete-previous-char and delete-next-char without an argument are not supposed to modify the yank buffer in Emacs. Make it so.
1.13	18-Oct-2003	christos	change allocation policy in el_push to allocate the string itself. fix issues with strdup.
1.12	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.11	20-Nov-2002	christos	Fix problem with previous patches that broke vi history. - c_gets() was usually returning a length, but sometimes one of the CC_xxx values (which are small +ve integers)! - fixed c_gets() by putting a ' ' under the cursor. From David Laight.
1.10	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.9	27-Oct-2002	christos	vi mode and memory fixes from david laight.
1.8	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.7	03-Feb-2002	christos	Fixed an __P remnant
1.6	10-Jan-2001	jdolecek	Enlarge editline buffers as needed to support arbitrary length lines. This also addresses lib/9712 by Phil Nelson.
1.5	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.4	02-Jul-1999	simonb	More trailing white space.
1.3	12-Dec-1998	christos	delint
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.17.28.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.21.22.1	10-Aug-2014	tls	Rebase.
1.21.12.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.37	10-Aug-2023	mrg	avoid various use-after-free issues. create a ptrdiff_t offset between the start of an allocation region and some interesting pointer, so it can be adjusted with this offset after realloc() returns. found by GCC 12.
1.36	30-Oct-2022	christos	improvements in malloc/free handling.
1.35	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.34	25-Nov-2018	christos	From Yuichiro Naito (FreeBSD): hrs@ says that wctomb(3) has an internal shift state, if wctomb(3) is called outside of libedit, the internal state can be changed and causes miscalculate multibyte size. So in this part, wcrtomb(3) should be used. wcrtomb(3) requires that shift state is given in the argument. We always initialize the shift state in ct_enc_width() to keep independent from outside of libedit.
1.33	18-Nov-2018	christos	fix compilation
1.32	18-Nov-2018	christos	Remove utf-8 requirement (Yuichiro NAITO)
1.31	09-Jan-2017	christos	branches: 1.31.12; 1.31.14; Make sure that argv is NULL terminated since functions like tty_stty rely on it to be so (Gerry Swinslow)
1.30	09-May-2016	christos	branches: 1.30.2; s/protected/libedit_private/g
1.29	02-May-2016	christos	eliminate static buffer with custom resizing code.
1.28	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.27	11-Apr-2016	christos	chartype cleanups from Ingo Schwarze: - The file tokenizer.c no longer uses chartype.h, so don't include the header. - The dummy definitions of ct_{de,en}code_string() for the NARROWCHAR case are only used in history.c, so move them there. - Now the whole content of chartype.h is for the wide character case only. So remove the NARROWCHAR ifdef and include the header only in the wide character case. - In chartype.h, move ct_encode_char() below the comment explaining it. - No more need for underscores before ct_{de,en}code_string(). - Make the conversion buffer resize functions private. They are only called from the decoding and encoding functions inside chartype.c, and no need can possibly arise to call them from anywhere else.
1.26	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.25	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.24	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.23	28-Feb-2016	christos	remove 4 clause licenses.
1.22	24-Feb-2016	christos	Tuck in mbstate_t to the wide char version only to avoid exposing the zeroing hack and doing it in the narrow case.
1.21	24-Feb-2016	christos	Make the read_char function always take a wchar_t * argument (Ingo Schwarze)
1.20	24-Feb-2016	christos	Get split el_getc and el_wgetc completely and call el_wgetc internally. Change some character constants to they wide versions. (Ingo Schwarze)
1.19	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.18	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.17	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.16	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.15	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.14	14-Feb-2016	christos	From Ingo Schwartze: Next step: Remove #ifdef'ing in read_char(), in the same style as we did for setlocale(3) in el.c. A few remarks are required to explain the choices made. * On first sight, handling mbrtowc(3) seems a bit less trivial than handling setlocale(3) because its prototype uses the data type mbstate_t from <wchar.h>. However, it turns out that "histedit.h" already includes <wchar.h> unconditionally (i don't like headers including other headers, but that ship has sailed, people are by now certainly used to the fact that including "histedit.h" doesn't require including <wchar.h> before), and "histedit.h" is of course included all over the place. So from that perspective, there is no problem with using mbrtowc(3) unconditionally ever for !WIDECHAR. * However, <wchar.h> also defines the mbrtowc(3) prototype, so we cannot just #define mbrtowc away, or including the header will break. It would also be a bad idea to porovide a local implementation of mbrtowc() and hope that it overrides the one in libc. Besides, the required prototype is subtly different: While mbrtowc(3) takes "wchar_t " as its first argument, we need a function that takes "Char ". So unfortunately, we have to keep a ct_mbrtowc #define, at least until we can maybe get rid of "Char " in the more remote future. After getting rid of the #else clause in read_char(), we can pull "return 1;" into the default: clause. After that, we can get rid of the ugly "goto again_lastbyte;" and just "break;". As a bonus, that also gets rid of the ugly CONSTCOND. * While here, delete the unused ct_mbtowc() from chartype.h.
1.13	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.12	22-Feb-2015	christos	split the allocation functions, their mixed usage was too confusing.
1.11	22-Feb-2015	christos	PR/49683: Amir Plivatsky: Off-by-one comparison in ct_decode_string() leading to out of bounds referrence. XXX: pullup-7
1.10	16-Aug-2011	christos	branches: 1.10.20; re-enable -Wconversion
1.9	29-Jul-2011	christos	pass -Wconversion
1.8	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.7	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.6	28-Jul-2011	christos	- fix memory allocation botch in wide strings - check mbstowcs return code
1.5	27-Jul-2011	christos	fix buffer growing code.
1.4	15-Apr-2010	christos	From Jess Thrysoee expose ct_enc_width()
1.3	12-Jan-2010	christos	- in the argv conversion, handle NULL as NULL - when printing tab/nl print them, don't handle them specially.
1.2	31-Dec-2009	christos	expose the encode and decode string functions for the benefit of history and readline.
1.1	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.10.20.1	14-Apr-2015	snj	Pull up following revision(s) (requested by christos in ticket #679): lib/libedit/chartype.c: revisions 1.11, 1.12 lib/libedit/chartype.h: revisions 1.12, 1.13 PR/49683: Amir Plivatsky: Off-by-one comparison in ct_decode_string() leading to out of bounds referrence. -- split the allocation functions, their mixed usage was too confusing.
1.30.2.1	20-Mar-2017	pgoyette	Sync with HEAD
1.31.14.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.31.14.1	10-Jun-2019	christos	Sync with HEAD
1.31.12.1	26-Nov-2018	pgoyette	Sync with HEAD, resolve a couple of conflicts
1.37	11-Apr-2022	tnn	libedit/chartype.h: portability fix for OSF/1
1.36	15-Sep-2019	christos	Fix type and remove cast (Yuichiro NAITO/FreeBSD).
1.35	22-May-2017	christos	branches: 1.35.10; Add DragonFly.
1.34	09-May-2016	christos	s/protected/libedit_private/g
1.33	02-May-2016	christos	eliminate static buffer with custom resizing code.
1.32	02-May-2016	christos	fix typos from Pedro Giffuni @FreeBSD
1.31	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.30	11-Apr-2016	christos	chartype cleanups from Ingo Schwarze: - The file tokenizer.c no longer uses chartype.h, so don't include the header. - The dummy definitions of ct_{de,en}code_string() for the NARROWCHAR case are only used in history.c, so move them there. - Now the whole content of chartype.h is for the wide character case only. So remove the NARROWCHAR ifdef and include the header only in the wide character case. - In chartype.h, move ct_encode_char() below the comment explaining it. - No more need for underscores before ct_{de,en}code_string(). - Make the conversion buffer resize functions private. They are only called from the decoding and encoding functions inside chartype.c, and no need can possibly arise to call them from anywhere else.
1.29	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.28	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.27	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.26	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.25	07-Mar-2016	christos	Remove advertising clause.
1.24	02-Mar-2016	christos	PR/50880: David Binderman: Remove redundant code. While here, fix all debugging formats.
1.23	24-Feb-2016	christos	Tuck in mbstate_t to the wide char version only to avoid exposing the zeroing hack and doing it in the narrow case.
1.22	24-Feb-2016	christos	Make the read_char function always take a wchar_t * argument (Ingo Schwarze)
1.21	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.20	14-Feb-2016	christos	From Ingo Schwarze: el_getc() for the WIDECHAR case, that is, the version in eln.c. For a UTF-8 locale, it is broken in four ways: 1. If the character read is outside the ASCII range, the function does an undefined cast from wchar_t to char. Even if wchar_t is internally represented as UCS-4, that is wrong and dangerous because characters beyond codepoint U+0255 get their high bits truncated, meaning that perfectly valid printable Unicode characters get mapped to arbitrary bytes, even the ASCII escape character for some Unicode characters. But wchar_t need not be implemented in terms of UCS-4, so the outcome of this function is undefined for any and all input. 2. If insufficient space is available for the result, the function fails to detect failure and returns garbage rather than -1 as specified in the documentation. 3. The documentation says that errno will be set on failure, but that doesn't happen either in the above case. 4. Even for ASCII characters, the results may be wrong if wchar_t is not using UCS-4.
1.19	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.18	14-Feb-2016	christos	From Ingo Schwartze: Next step: Remove #ifdef'ing in read_char(), in the same style as we did for setlocale(3) in el.c. A few remarks are required to explain the choices made. * On first sight, handling mbrtowc(3) seems a bit less trivial than handling setlocale(3) because its prototype uses the data type mbstate_t from <wchar.h>. However, it turns out that "histedit.h" already includes <wchar.h> unconditionally (i don't like headers including other headers, but that ship has sailed, people are by now certainly used to the fact that including "histedit.h" doesn't require including <wchar.h> before), and "histedit.h" is of course included all over the place. So from that perspective, there is no problem with using mbrtowc(3) unconditionally ever for !WIDECHAR. * However, <wchar.h> also defines the mbrtowc(3) prototype, so we cannot just #define mbrtowc away, or including the header will break. It would also be a bad idea to porovide a local implementation of mbrtowc() and hope that it overrides the one in libc. Besides, the required prototype is subtly different: While mbrtowc(3) takes "wchar_t " as its first argument, we need a function that takes "Char ". So unfortunately, we have to keep a ct_mbrtowc #define, at least until we can maybe get rid of "Char " in the more remote future. After getting rid of the #else clause in read_char(), we can pull "return 1;" into the default: clause. After that, we can get rid of the ugly "goto again_lastbyte;" and just "break;". As a bonus, that also gets rid of the ugly CONSTCOND. * While here, delete the unused ct_mbtowc() from chartype.h.
1.17	11-Feb-2016	christos	remove unused wrapper (Ingo Schwarze)
1.16	08-Feb-2016	christos	UTF-8 fixes from Ingo Schwarze: 1. Assume that errno is non-zero when entering read_char() and that read(2) returns 0 (indicating end of file). Then, the code will clear errno before returning. (Obviously, the statement "errno = 0" is almost always a bug unless there is save_errno = errno right before it and the previous value is properly restored later, in all reachable code paths.) 2. When encountering an invalid byte sequence, the code discards all following bytes until MB_LEN_MAX overflows; consider, for example, 0xc2 immediately followed by a few valid ASCII bytes. Three of those ASCII bytes will be discarded. 3. On a POSIX system, EILSEQ will always be set after reading a valid (yes, valid, not invalid!) UTF-8 character. The reason is that mbtowc(3) will first be called with a length limit (third argument) of 1, which will fail, return -1, and - on a POSIX system - set errno to EILSEQ. This third bug is mitigated a bit because i couldn't find any system that actually conforms to POSIX in this respect: None of OpenBSD, NetBSD, FreeBSD, Solaris 11, and glibc set errno when an incomplete character is passed to mbtowc(3), even though that is required by POSIX. Anyway, that mbtowc(3) bug will be fixed at least in OpenBSD after release unlock, so it would be good to fix this bug in libedit before fixing the bug in mbtowc(3). How can these three bugs be fixed? 1. As far as i understand it, the intention of the bogus errno = 0 is to undo the effects of failing system calls in el_wset(), sig_set(), and read__fixio() if the subsequent read(2) indicates end of file. So, restoring errno has to be moved right after read__fixio(). Of course, neither 0 nor e is the right value to restore: 0 is wrong if errno happened to be set on entry, e would be wrong because if one read(2) fails but a second attempt succeeds after read__fixio(), errno should not be touched. So, the errno to be restored in this case has to be saved before calling read(2) for the first time. 2. Solving the second issue requires distinguishing invalid and incomplete characters, but that is impossible with the function mbtowc(3) because it returns -1 in both cases and sets errno to EILSEQ in both cases (once properly implemented). It is vital that each input character is processed right away. It is not acceptable to wait for the next input character before processing the previous one because this is an interactive library, not a batch system. Consequently, the only situation where it is acceptable to wait for the next byte without first processing the previous one(s) is when the previous one(s) form an incomplete sequence that can be continued to form a valid character. Consequently, short of reimplementing a full UTF-8 state machine by hand, the only correct way forward is to use mbrtowc(3). Even then, care is needed to always have the state object properly initialized before using it, and to not discard a valid ASCII or UTF-8 lead byte if it happens to follow an invalid sequence. 3. Fortunately, solution 2. also solves issue 3. as a side effect, by no longer using mbtowc(3) in the first place.
1.15	17-May-2015	christos	add FreeBSD
1.14	14-May-2015	christos	fix warnings on ubuntu 32 bit (Miki Rozloznik)
1.13	22-Feb-2015	christos	split the allocation functions, their mixed usage was too confusing.
1.12	22-Feb-2015	christos	PR/49683: Amir Plivatsky: Off-by-one comparison in ct_decode_string() leading to out of bounds referrence. XXX: pullup-7
1.11	17-Feb-2015	christos	OpenBSD is like us.
1.10	16-Nov-2011	christos	branches: 1.10.18; easier with an int for now.
1.9	15-Nov-2011	christos	Since Width() is used only for display purposes we don't want to pass -1 for unprintable characters.
1.8	29-Jul-2011	christos	branches: 1.8.2; pass -Wconversion
1.7	16-Dec-2010	wiz	Observe the following spelling: - wide character (noun) - wide-character (adjective) Inspired by jmc@OpenBSD.
1.6	20-Apr-2010	christos	Use the same hack for Solaris and MacOS/X. This is not right, we only really support UTF-8, but it will get us going until this is fixed properly. From Jess Thrysoee
1.5	15-Apr-2010	christos	From Jess Thrysoee expose ct_enc_width()
1.4	03-Jan-2010	christos	rename historyw -> history_w for consistency. add wide tst code and make it the default.
1.3	31-Dec-2009	christos	expose the encode and decode string functions for the benefit of history and readline.
1.2	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.1	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.8.2.1	17-Apr-2012	yamt	sync with head
1.10.18.2	13-May-2015	martin	Sync lib/libedit with head, requested by christos in #753: lib/libedit/Makefile 1.53 lib/libedit/chartype.h 1.13 lib/libedit/editline.3 1.83-1.84 lib/libedit/editrc.5 1.28-1.29 lib/libedit/eln.c 1.18 lib/libedit/filecomplete.c 1.33-1.34 lib/libedit/readline.c 1.112-1.115 Man page improvements, fix overlapping strcpy, improve readline compatibility, clang build fix.
1.10.18.1	14-Apr-2015	snj	Pull up following revision(s) (requested by christos in ticket #679): lib/libedit/chartype.c: revisions 1.11, 1.12 lib/libedit/chartype.h: revisions 1.12, 1.13 PR/49683: Amir Plivatsky: Off-by-one comparison in ct_decode_string() leading to out of bounds referrence. -- split the allocation functions, their mixed usage was too confusing.
1.35.10.1	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.50	30-Jun-2024	christos	Avoid moving the cursor before the buffer (Robert Morris) https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=279487
1.49	30-Mar-2020	ryo	branches: 1.49.8; fix build error with SDEBUG, MAP_DEBUG, DEBUG_REFRESH
1.48	26-Feb-2018	christos	branches: 1.48.4; PR/53058: Nikhil Benesch: use correctly typed variables (wchar_t vs wint_t) as parameters.
1.47	22-May-2016	christos	Stop the read module from poking the el_chared.c_macro data structure currently belonging to the chared module. The read module does so from three of its functions, while no other module uses the macro data, not even the chared module itself. That's quite logical because macros are a feature of input handling, all of which is done by the read module, and none by the chared module. So move the data into the read modules's own opaque data structure, struct el_read_t. That simplifies internal interfaces in several respects: The semi-public chared.h has one fewer struct, one fewer #define, and one fewer member in struct el_chared_t; all three move to one single C file, read.c, and are now module-local. And the internal interface function ch_reset() needs one fewer argument, making the code of many functions in various modules more readable. The price is one additional internal interface function, read_end(), 10 lines long including comments, called publicly from exactly one place: el_end() in el.c. That's hardly an increase in complexity since most other modules already have their *_end() function, read.c was the odd one out not having one. From Ingo Schwarze
1.46	09-May-2016	christos	s/protected/libedit_private/g
1.45	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.44	17-Apr-2016	christos	Remove empty callbacks (Ingo Schwartze)
1.43	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.42	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.41	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.40	02-Mar-2016	christos	PR/50880: David Binderman: Remove redundant code. While here, fix all debugging formats.
1.39	24-Feb-2016	christos	Get split el_getc and el_wgetc completely and call el_wgetc internally. Change some character constants to they wide versions. (Ingo Schwarze)
1.38	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.37	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.36	16-Feb-2016	christos	get rid of bool_t (Ingo Schwarze)
1.35	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.34	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.33	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.32	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.31	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.30	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.29	24-Mar-2012	christos	From Jilles Tjoelker: Do not move the cursor for ed-delete-next-char in emacs mode. This makes ed-delete-next-char suitable for mapping to the <Delete> key. Behaviour in vi mode is unchanged (for 'x').
1.28	29-Jul-2011	christos	branches: 1.28.2; - fix unused params - unconditionalize vis.h
1.27	29-Jul-2011	christos	kill ifdef notdef
1.26	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.25	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.24	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.23	27-Feb-2009	msaitoh	fix mis-evaluating whether a char is digit or not.
1.22	15-Feb-2009	christos	pass lint on _LP64.
1.21	30-Sep-2008	aymeric	branches: 1.21.4; 1.21.6; 1.21.10; have '$' include the last character in the line when embedded in a command. This fixes c$, d$, y$, and so on in vi mode.
1.20	10-Sep-2008	christos	Allow a single process to control multiple ttys (for pthreads using _REENTRANT) using multiple EditLine objects. Mostly from Preston A. Elder.
1.19	06-Mar-2006	christos	branches: 1.19.4; 1.19.20; Print the actual eofc, instead of ^D\b\b. Change internal character decoding to prevent buffer oveflows.
1.18	08-Aug-2005	christos	Spelling mistakes and comment errors (from FreeBSD via Stefan Farfeleder; many thanks)
1.17	01-Aug-2005	christos	Don't reset the macro strings each time we enter el_gets(), otherwise el_push() is unusable programmatically.
1.16	07-Aug-2003	agc	branches: 1.16.6; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.15	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.14	20-Nov-2002	christos	Fix problem with previous patches that broke vi history. - c_gets() was usually returning a length, but sometimes one of the CC_xxx values (which are small +ve integers)! - fixed c_gets() by putting a ' ' under the cursor. From David Laight.
1.13	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.12	27-Oct-2002	christos	vi mode and memory fixes from david laight.
1.11	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.10	10-Jan-2001	jdolecek	Enlarge editline buffers as needed to support arbitrary length lines. This also addresses lib/9712 by Phil Nelson.
1.9	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.8	02-Jul-1999	simonb	More trailing white space.
1.7	12-Dec-1998	christos	delint
1.6	20-May-1998	christos	Don't print to stderr, but to the editline error stream.
1.5	03-Feb-1998	perry	remove obsolete register declarations
1.4	06-Jul-1997	christos	Fix compiler warnings.
1.3	14-Jan-1997	lukem	Implement CC_REDISPLAY, which (unlink CC_REFRESH) redraws the entire input line (a la ^R). This is useful if the binding outputs information and mucks up the input line. To be used in ``list-choices'' bindings (refer to the ^D binding in csh when filec is set)
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.16.6.1	23-Apr-2009	snj	Pull up following revision(s) (requested by msaitoh in ticket #2006): lib/libedit/common.c: revision 1.23 fix mis-evaluating whether a char is digit or not.
1.19.20.1	24-Sep-2008	wrstuden	Merge in changes between wrstuden-revivesa-base-2 and wrstuden-revivesa-base-3.
1.19.4.1	30-Apr-2009	bouyer	Pull up following revision(s) (requested by msaitoh in ticket #1307): lib/libedit/common.c: revision 1.23 fix mis-evaluating whether a char is digit or not.
1.21.10.1	21-Apr-2010	matt	sync to netbsd-5
1.21.6.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.21.4.1	03-May-2009	snj	Pull up following revision(s) (requested by msaitoh in ticket #703): lib/libedit/common.c: revision 1.23 fix mis-evaluating whether a char is digit or not.
1.28.2.1	17-Apr-2012	yamt	sync with head
1.48.4.1	08-Apr-2020	martin	Merge changes from current as of 20200406
1.49.8.1	01-Jul-2024	perseant	Sync with HEAD.
1.10	04-Feb-2023	christos	Remove unused stuff, and limit the scope of some of the used ones. (from des@freebsd)
1.9	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.8	16-Feb-2016	christos	- don't set _GNU_SOURCE. We are not supposed to make decisions for others. - don't special-case wcsdup() From Ingo Schwarze.
1.7	15-Feb-2016	christos	OpenBSD readline.c rev. 1.14 2015/02/06 23:21:58 millert use SIZE_MAX
1.6	15-Feb-2016	christos	change tests for fgetln.
1.5	29-Jul-2011	christos	better autoconf results
1.4	28-Jul-2011	joerg	Sync compat defines to unbreak build. XXX Do a proper config.h update from configure
1.3	28-Jul-2011	christos	regen
1.2	26-Jan-2009	apb	Define HAVE_STRUCT_DIRENT_D_NAMLEN in config,h, and test it when deciding whether to use entry->d_namlen or strlen(entry->d_name). Addresses PR 40477 by Robert Millan.
1.1	18-Mar-2002	christos	oops, we need config.h now.
1.1	21-Nov-2024	riastradh	branches: 1.1.4; libedit: Add expected symbols list. PR lib/58838: shared libraries in base should all have expsym lists
1.1.4.2	02-Aug-2025	perseant	Sync with HEAD
1.1.4.1	21-Nov-2024	perseant	file edit.expsym was added on branch perseant-exfatfs on 2025-08-02 05:54:48 +0000
1.102	04-Feb-2024	andvar	s/interrrupt/interrupt/ in comment and editline(3) man page.
1.101	15-Aug-2021	wiz	Add verb to sentence.
1.100	15-Aug-2021	christos	Disable attempts to handle EINTR and non-blocking I/O by default. It is confusing to other programs and unexpected behavior. Reported by Ingo Schwarze. This behavior is now controlled with EL_SAFEREAD.
1.99	18-Nov-2018	christos	Remove utf-8 requirement (Yuichiro NAITO)
1.98	02-Sep-2017	wiz	branches: 1.98.2; 1.98.4; Remove trailing whitespace; convert Xr readline to Nm readline, since we won't get a readline man page in base.
1.97	01-Sep-2017	christos	PR/51517: Jay West: Tty settings not restored on exit PR/51518: Jay West: prompt is interleaved with client output Both these issues are caused by rl_restore_handler not DTRT; fix it so that it kills the internal libedit state completely. This is inefficient, but it works. Also fix: 1. add append_history()/H_NSAVE_FP 2. call the rl_startup_hook before printing the first prompt as documented. callint it from rl_initialize breaks python, because the callback ends up being invoked before the readline module is installed, and we end up dereferencing a NULL pointer. 3. add el_resize_terminal. With those changes, s/lreadline/ledit/g in python works.
1.96	03-Jul-2017	wiz	Remove workaround for ancient HTML generation code.
1.95	27-Jun-2017	christos	mention the limitation of the literal sequence delimiter.
1.94	27-Jun-2017	kre	Include EDITRC in doc.
1.93	10-Apr-2017	abhinav	branches: 1.93.4; Add missing argument for H_SET. ok christos@
1.92	22-May-2016	christos	branches: 1.92.2; 1.92.4; documentation improvements (Ingo Schwarze)
1.91	21-May-2016	christos	Fix the prototype used by EL_GETCFN, mention the associated typedef name, document the return values, expand the list of affected functions, warn against using EL_GETCFN, and clarify some wording and notation. (Ingo Schwarze)
1.90	09-May-2016	christos	GNU readline(3) regards history chronologically, that is, from the perspective of the dawn of time, so "next" means "newer" and "previous" means "older". Libedit, by contrast, uses reverse chronology and regards history from the perspective of the present, such that "next" means "longer ago" and "previous" means "not so long ago". The following patch fixes previous_history() and next_history() as proposed by Bastian Maerkisch. But there is a related problem demonstrated by Bastian's regression tests that his patch did not fix: next_history() can advance not only to the newest entry, but beyond it, which core libedit cannot do. So that feature must be implemented locally in readline.c. With that, the last of Bastians tests is fixed, test_movement_direction(). This patch also improves libedit documentation to more clearly state what "previous" and "next" mean. GNU readline documentation is just as unclear, but we can't easily fix that since libedit doesn't include its own readline.3 manual. (Ingo Schwarze)
1.89	28-Apr-2016	christos	new man page from Ingo Schwarze.
1.88	25-Feb-2016	wiz	Use \- for minus sign, use Ev, use Er.
1.87	24-Feb-2016	christos	Fix el_{w,}getc documentation (Ingo Schwarze)
1.86	24-Feb-2016	christos	Fixes from OpenBSD via Ingo Schwarze: 1) Missing comma after tok_str in NAME. OpenBSD rev. 1.38 Sep 10, 2015 (schwarze) 2) Style: void in argument list. OpenBSD rev. 1.39 Sep 14, 2015 (schwarze) 3) English punctuation: stray comma. OpenBSD rev. 1.37 Mar 13, 2015 (jmc)
1.85	03-Nov-2015	christos	Fix descriptions of el_set functions. Americanise initialise :-)
1.84	25-Dec-2014	wiz	Bump date for previous.
1.83	25-Dec-2014	wiz	From Ingo Schwarze, based on changes from Kaspars Bankovskis: * Document error handling of el_init(), el_set(), el_source(), and history_init(). * Fix a typo an improve punctuation below H_SETUNIQUE. * The ellipsis already implies "optional", no need for []. * Sort options in editrc(5). * Prevent e.g. rom being misconstrued as the end of a sentence. * Drop a useless duplicate .Ar macro. * Put telltc in its proper place in the alphabetical order. * A few typos in vi editor command names. * Some missing vi editor command names. * Some missing author macros.
1.82	11-May-2014	wiz	branches: 1.82.2; Add An to authors. Wording.
1.81	11-May-2014	christos	Add a history function that takes a FILE pointer; needed for Capsicum. From Eitan Adler
1.80	12-Jul-2013	christos	branches: 1.80.4; Add a function to move the cursor.
1.79	22-Jan-2013	christos	provide an el_init_fd function.
1.78	10-Jan-2013	wiz	In 2000, .editrc reading from $PWD was removed. Update the man page. From LEVAI Daniel via jmc@OpenBSD.
1.77	11-Sep-2012	christos	branches: 1.77.2; PR/46945: Steffen Nurpmeso; el_getc() doesn't document it's setting errno
1.76	11-Sep-2012	christos	PR/46941: Steffen Nurpmeso: document EL_BUFFERED
1.75	27-Feb-2011	christos	branches: 1.75.4; Fix argument for EL_EDITOR from Jess Thrysoee
1.74	16-Dec-2010	wiz	branches: 1.74.2; Observe the following spelling: - wide character (noun) - wide-character (adjective) Inspired by jmc@OpenBSD.
1.73	03-Jan-2010	wiz	Bump date for historyw -> history_w.
1.72	03-Jan-2010	christos	rename historyw -> history_w for consistency. add wide tst code and make it the default.
1.71	31-Dec-2009	christos	- Document and enable wide character support. - Fix read function compatibility.
1.70	05-Jul-2009	perry	note that the return value of el_gets doesn't remain valid across calls.
1.69	12-May-2009	wiz	Punctuation nit.
1.68	11-May-2009	christos	restore binary compatibility by providing new prompt functions that take an extra literal character.
1.67	11-Apr-2009	joerg	Don't use .Xo/.Xc to avoid ancient macro argument limit.
1.66	01-Apr-2009	wiz	Readability improvement. Whitespace nits.
1.65	31-Mar-2009	christos	Implement literal prompt sequences. Now someone can implement RL_PROMPT_START_LITERAL/RL_PROMPT_END_LITERAL :-)
1.64	10-Mar-2009	wiz	-1 is not a defined value, it is just a value. On the other hand, minuses need backslashes, otherwise they become hyphens.
1.63	10-Mar-2009	christos	make el_gets set the count to -1 on error to distinguish between EOF and error.
1.62	09-Mar-2009	joerg	Fix preamble to match order set out by mdoc(7). Discussed with wiz.
1.61	21-Feb-2009	wiz	Restore markup changes (probably accidentally) reverted in previous.
1.60	21-Feb-2009	christos	back out all prompt changes. they are not needed.
1.59	19-Feb-2009	wiz	Fix wordo, use more markup.
1.58	19-Feb-2009	christos	document extra argument.
1.57	30-Apr-2008	martin	branches: 1.57.8; Convert TNF licenses to new 2 clause variant
1.56	05-Apr-2008	christos	branches: 1.56.2; add EL_REFRESH for the benefit of readline
1.55	12-Jan-2007	christos	PR/35411: Matthew Wala: inconsistency in editline(3): rename "num" to the appropriate parameter names.
1.54	18-Dec-2006	wiz	Fix xref section. Whitespace cleanups..
1.53	15-Dec-2006	christos	add EL_GETFP, and EL_SETFP.
1.52	24-Nov-2006	christos	- Add more readline functions, enough for gdb-6.5 - Make el_get varyadic, and implement EL_GETTC. - XXX: the EL_SETTC api will change in the future.
1.51	21-Aug-2006	christos	Change to a 3 clause copyright after permission of the holders.
1.50	12-Oct-2005	wiz	Add el_get to SYNOPSIS. From jmc@openbsd.
1.49	09-Sep-2005	wiz	Fix double if (from Alexey E. Suslikov via jmc@openbsd). While here, re-word both H_[GS]ETUNIQUE descriptions so they make more sense. Bump date.
1.48	14-Jul-2005	wiz	Drop trailing whitespace.
1.47	14-Jul-2005	christos	PR/30747: David N. Williams: libedit is missing remove_history() Added, please test.
1.46	19-Mar-2005	christos	PR/29738: Rui Paulo: Incorrect tok_line and tok_str declarations.
1.45	15-Apr-2004	wiz	Remove duplicate word, from jfb@openbsd via jmc@openbsd.
1.44	05-Dec-2003	wiz	Use Aq instead of <>, quote a minus, drop trailing space.
1.43	05-Dec-2003	lukem	Tokenization function enhancements: * Make tok_init(), tok_end(), tok_reset(), tok_line() and tok_str() publically available in <histedit.h> * Documented the public functions in editline(3) * Renamed tok_line() -> tok_str() * Added new tok_line() which takes a "const LineInfo " instead of "const char " (the former has "cursor" information), and optionally return the argv index ("int cursorc") and offset within that index ("int cursorv"). This means that completion routines can use the tokenization code to crack the line and easily find which word the cursor is at. (mmm, context sensitive completion :) * Fixed TEST/test.c when using "continuation" lines (unmatched quote or \ at EOL), and added some more DEBUG messages including highlighting where the cursor is (with a `_').
1.42	04-Nov-2003	christos	s/wth/with/
1.41	01-Nov-2003	christos	Explain H_ADD better. from Otto Moerbeek otto at drijf dot net
1.40	17-Oct-2003	wiz	Bump date for previous. Replace > with \*[Gt].
1.39	16-Oct-2003	christos	More libedit readline emulation functions from: Gerry Swislow <gerry at certif dot com>
1.38	26-Sep-2003	wiz	New sentence, new line; bump date for previous.
1.37	26-Sep-2003	christos	Implement enough of readline's 4.0 async mode to make gdb happy. This is not complete yet, but it seems to work... This required to introduce an unbuffered mode to el_gets(), but that was a minor change.
1.36	16-Jul-2003	wiz	Bump date for last.
1.35	14-Jul-2003	jeremy	Documented the return values from el_get().
1.34	06-Jun-2003	wiz	Use Aq Pa instead of Fd \[Lt]...\[Gt]. From jmc@openbsd.
1.33	16-Apr-2003	wiz	Use .In header.h instead of .Fd #include \[Lt]header.h\[Gt] Much easier to read and write, and supported by groff for ages. Okayed by ross.
1.32	25-Feb-2003	wiz	.Nm does not need a dummy argument ("") before punctuation or for correct formatting of the SYNOPSIS any longer.
1.31	21-Jan-2003	wiz	Replace -1 with \-1 for PostScript output; drop a trailing space and fix two typos.
1.30	21-Jan-2003	christos	Add a uniquefier for the history function.
1.29	01-Oct-2002	wiz	New sentence, new line. From Robert Elz.
1.28	18-Aug-2002	yamt	reflect reality. (3rd arg of el_parse is const.)
1.27	20-Feb-2002	wiz	Close quoting.
1.26	07-Feb-2002	ross	Generate <>& symbolically.
1.25	15-Jan-2002	wiz	Punctuation and whitespace nits, fix a typo.
1.24	09-Oct-2001	christos	PR/14188: Anthony Mallet: Provide an opaque data pointer to client programs.
1.23	04-Oct-2001	lukem	a couple of minor fixes. originally by Ruslan Ermilov <ru@FreeBSD.org>, highlighted to me by way of Mike Barcroft <mike@FreeBSD.org> (thanks!)
1.22	27-Sep-2001	christos	PR/14067: Anthony Mallet: Provide a programmatic way to set the read_char function via a new el_set() operation. Thanks, nicely done :-)
1.21	02-Apr-2001	wiz	End sentence with a dot.
1.20	28-Feb-2000	chopps	el_insertstr takes a "const char " not "char " now as it doesn't modify the argument.
1.19	26-Nov-1999	lukem	missing ,
1.18	12-Nov-1999	lukem	- implement printing a right-side prompt. code derived from similar work I wrote for tcsh(1) three years ago. - implement EL_RPROMPT, which allows a setting/getting of a function which returns a string to be used as the right-side prompt. - improve HISTORY and AUTHORS sections in editline(3). - bump shlib minor version for EL_RPROMPT. XXX: due to an implementation issue, the rprompt has a 1 space gap before the edge of the logical screen. editline's logical screen is 1 space less than the full screen width, so there's a 2 space gap between the rprompt and the right end of the physical screen. i'm not concerned about this.
1.17	02-Jul-1999	simonb	branches: 1.17.4; More trailing white space.
1.16	22-Mar-1999	garbled	Last of the .Os cleanups. .Os is defined in the tmac.doc-common file, so we shouldn't override it with versions in the manpages. Wheee!
1.15	05-Jan-1999	lukem	fix history() prototype. (d@openbsd.org)
1.14	29-Jul-1998	lukem	* add more checks for NULL pointers in passed arguments * implement el_get(EditLine , int op, void result), which does the inverse of el_set() * add EL_EDITMODE operation to el_set and el_get; if non zero editing is enabled (the default). * add "edit on \| off" editrc command, which modifies EL_EDITMODE. users can now add '*:edit off' in ~/.editrc as an advisory to disable editing. NOTE: at this time EL_EDITMODE is just an indication of the state of the 'edit' command. It's up to the application to check this after el_source() or el_parse() to determine if editing is still required.
1.13	08-Jun-1998	lukem	documentation is always probably incomplete; don't make an issue of it
1.12	01-Jun-1998	lukem	* implement CC_REFRESH_BEEP; as per CC_REFRESH but beep as well. this is useful in completion when a partial completion is found * remove entry in BUGS about el_parse(); that was fixed a while ago
1.11	20-May-1998	christos	Add H_APPEND to simplify the interface.
1.10	20-May-1998	christos	Adjust for changes.
1.9	05-Feb-1998	perry	add LIBRARY section to man page
1.8	21-Jan-1998	lukem	in el_parse(), use a temporary buffer to store the program name when comparing, preventing trashing of argv[0]. remove note in man page warning of former behaviour.
1.7	09-Nov-1997	lukem	move description of history() return value
1.6	14-Oct-1997	christos	PR/4257: Jaromir Dolecek: history() has no generic error handling and isn't reentrant. This changes the interface of the history function, so we need a major number bump.
1.5	30-Jul-1997	jtc	Fix files using old TNF copyright notice
1.4	14-Jan-1997	lukem	Implement CC_REDISPLAY, which (unlink CC_REFRESH) redraws the entire input line (a la ^R). This is useful if the binding outputs information and mucks up the input line. To be used in ``list-choices'' bindings (refer to the ^D binding in csh when filec is set)
1.3	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.2	11-Jan-1997	lukem	* in el_parse(), don't reference argv[0] if argc < 1 (return -1 instead) * clarify return value of el_parse()
1.1	09-Jan-1997	lukem	* add a man page for the editline routines * add a man page describing editrc * fix bugs in el_parse(): * didn't execute command when program name matched (test reversed) * was checking against empty string instead of program name * after checks, command to run also pointed to empty string [christos - the author of libedit - ok-ed the man pages in general (which I wrote from scratch by RTFS) as well as the bugfix]
1.17.4.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.56.2.1	18-May-2008	yamt	sync with head.
1.57.8.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.74.2.1	05-Mar-2011	bouyer	Sync with HEAD
1.75.4.3	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.75.4.2	23-Jan-2013	yamt	sync with head
1.75.4.1	30-Oct-2012	yamt	sync with head
1.77.2.2	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.77.2.1	25-Feb-2013	tls	resync with head
1.80.4.1	10-Aug-2014	tls	Rebase.
1.82.2.1	13-May-2015	martin	Sync lib/libedit with head, requested by christos in #753: lib/libedit/Makefile 1.53 lib/libedit/chartype.h 1.13 lib/libedit/editline.3 1.83-1.84 lib/libedit/editrc.5 1.28-1.29 lib/libedit/eln.c 1.18 lib/libedit/filecomplete.c 1.33-1.34 lib/libedit/readline.c 1.112-1.115 Man page improvements, fix overlapping strcpy, improve readline compatibility, clang build fix.
1.92.4.1	21-Apr-2017	bouyer	Sync with HEAD
1.92.2.1	26-Apr-2017	pgoyette	Sync with HEAD
1.93.4.1	23-Jul-2017	snj	Pull up following revision(s) (requested by kre in ticket #102): lib/libedit/Makefile: 1.64-1.65 lib/libedit/editline.3: 1.94-1.96 lib/libedit/editrc.5: 1.33 lib/libedit/el.c: 1.93-1.94 lib/libedit/el.h: 1.42 lib/libedit/literal.c: 1.1-1.3 lib/libedit/literal.h: 1.1-1.2 lib/libedit/prompt.c: 1.27 lib/libedit/read.c: 1.103 lib/libedit/refresh.c: 1.52-1.54 lib/libedit/refresh.h: 1.11 lib/libedit/terminal.c: 1.33 Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming. -- Include EDITRC in doc. -- mention the limitation of the literal sequence delimiter. -- - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display. -- - add literal sequence handling. -- remove unused variable -- add literal escape sequence support, patterned after the tcsh ones. -- fix comment -- Fix an obvious, but almost invisible typo (avoid some core dumps). -- Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale). -- Remove workaround for ancient HTML generation code.
1.98.4.1	10-Jun-2019	christos	Sync with HEAD
1.98.2.1	26-Nov-2018	pgoyette	Sync with HEAD, resolve a couple of conflicts
1.6	06-Apr-2024	christos	update em-toggle-overwrite binding (Xose Vazquez Perez)
1.5	09-May-2016	christos	GNU readline(3) regards history chronologically, that is, from the perspective of the dawn of time, so "next" means "newer" and "previous" means "older". Libedit, by contrast, uses reverse chronology and regards history from the perspective of the present, such that "next" means "longer ago" and "previous" means "not so long ago". The following patch fixes previous_history() and next_history() as proposed by Bastian Maerkisch. But there is a related problem demonstrated by Bastian's regression tests that his patch did not fix: next_history() can advance not only to the newest entry, but beyond it, which core libedit cannot do. So that feature must be implemented locally in readline.c. With that, the last of Bastians tests is fixed, test_movement_direction(). This patch also improves libedit documentation to more clearly state what "previous" and "next" mean. GNU readline documentation is just as unclear, but we can't easily fix that since libedit doesn't include its own readline.3 manual. (Ingo Schwarze)
1.4	02-May-2016	wiz	Fix Dd argument.
1.3	02-May-2016	christos	Add more explicit xrefs
1.2	02-May-2016	wiz	Fix Dd argument.
1.1	28-Apr-2016	christos	new man page from Ingo Schwarze.
1.2	09-May-2016	christos	Instead of compiling all the source files together in one big file, use protected visibility to achieve the same effect.
1.1	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.34	06-Dec-2022	uwe	editrc(5): Markup fixes
1.33	27-Jun-2017	kre	Include EDITRC in doc.
1.32	22-May-2016	christos	branches: 1.32.8; documentation improvements (Ingo Schwarze)
1.31	28-Apr-2016	christos	new man page from Ingo Schwarze.
1.30	17-Apr-2016	christos	Remove empty callbacks (Ingo Schwartze)
1.29	25-Dec-2014	wiz	Bump date for previous.
1.28	25-Dec-2014	wiz	From Ingo Schwarze, based on changes from Kaspars Bankovskis: * Document error handling of el_init(), el_set(), el_source(), and history_init(). * Fix a typo an improve punctuation below H_SETUNIQUE. * The ellipsis already implies "optional", no need for []. * Sort options in editrc(5). * Prevent e.g. rom being misconstrued as the end of a sentence. * Drop a useless duplicate .Ar macro. * Put telltc in its proper place in the alphabetical order. * A few typos in vi editor command names. * Some missing vi editor command names. * Some missing author macros.
1.27	10-Jan-2013	wiz	branches: 1.27.8; Add FILES section. From jmc@OpenBSD.
1.26	02-Jun-2012	njoly	branches: 1.26.2; Switch from Op to Oo/Oc for nested block.
1.25	25-Apr-2011	wiz	branches: 1.25.4; Markup consistency, from YOMURA Masanori. Add serial comma.
1.24	11-Apr-2009	wiz	Drop trailing whitespace.
1.23	11-Apr-2009	joerg	Don't use .Xo/.Xc to work around ancient macro argument limit in groff.
1.22	09-Mar-2009	joerg	Fix preamble to match order set out by mdoc(7). Discussed with wiz.
1.21	30-Apr-2008	martin	branches: 1.21.8; Convert TNF licenses to new 2 clause variant
1.20	21-Aug-2006	christos	branches: 1.20.16; Change to a 3 clause copyright after permission of the holders.
1.19	01-Nov-2003	christos	Document history builtin commands.
1.18	19-Oct-2003	wiz	Bump date for previous.
1.17	18-Oct-2003	christos	Allow setty to set chars using char=value
1.16	27-Jun-2003	wiz	Add Ns.
1.15	14-May-2003	wiz	setup -> set up, from jmc@openbsd.
1.14	31-Mar-2003	perry	em-gosmacs-traspose->em-gosmacs-transpose (Igor Sobrado, PR misc/19909)
1.13	02-Oct-2002	wiz	empty, not emptry. By Adrian Mrva.
1.12	15-Jan-2002	wiz	Punctuation nits.
1.11	19-Jun-2001	wiz	`existent', not `existant'
1.10	08-Nov-2000	lukem	improve description of line syntax
1.9	21-Jun-2000	lukem	* add -m option to makelist, which generates an mdoc table with the key bindings and their descriptions * manually add the output of 'sh ./makelist -m vi.c ed.c common.c' to a new section in editrc(5) called `EDITOR COMMANDS'
1.8	02-Jul-1999	simonb	branches: 1.8.8; More trailing white space.
1.7	22-Mar-1999	garbled	Last of the .Os cleanups. .Os is defined in the tmac.doc-common file, so we shouldn't override it with versions in the manpages. Wheee!
1.6	29-Jul-1998	lukem	* add more checks for NULL pointers in passed arguments * implement el_get(EditLine , int op, void result), which does the inverse of el_set() * add EL_EDITMODE operation to el_set and el_get; if non zero editing is enabled (the default). * add "edit on \| off" editrc command, which modifies EL_EDITMODE. users can now add '*:edit off' in ~/.editrc as an advisory to disable editing. NOTE: at this time EL_EDITMODE is just an indication of the state of the 'edit' command. It's up to the application to check this after el_source() or el_parse() to determine if editing is still required.
1.5	30-Jul-1997	jtc	Fix files using old TNF copyright notice
1.4	24-Apr-1997	christos	editrc -> editline
1.3	11-Jan-1997	lukem	* document ^char and \ escape sequences * when parsing ^char control chars, check the correct char when determining validity (previously, ^char was a NOP interpreted as the literal string because of this bug)
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	09-Jan-1997	lukem	* add a man page for the editline routines * add a man page describing editrc * fix bugs in el_parse(): * didn't execute command when program name matched (test reversed) * was checking against empty string instead of program name * after checks, command to run also pointed to empty string [christos - the author of libedit - ok-ed the man pages in general (which I wrote from scratch by RTFS) as well as the bugfix]
1.8.8.1	23-Jun-2000	lukem	Pull up editrc.5 revision 1.9 Pull up makelist revision 1.5 * add -m option to makelist, which generates an mdoc table with the key bindings and their descriptions * manually add the output of 'sh ./makelist -m vi.c ed.c common.c' to a new section in editrc(5) called `EDITOR COMMANDS' requested/approved by thorpej
1.20.16.1	18-May-2008	yamt	sync with head.
1.21.8.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.25.4.2	23-Jan-2013	yamt	sync with head
1.25.4.1	30-Oct-2012	yamt	sync with head
1.26.2.1	25-Feb-2013	tls	resync with head
1.27.8.1	13-May-2015	martin	Sync lib/libedit with head, requested by christos in #753: lib/libedit/Makefile 1.53 lib/libedit/chartype.h 1.13 lib/libedit/editline.3 1.83-1.84 lib/libedit/editrc.5 1.28-1.29 lib/libedit/eln.c 1.18 lib/libedit/filecomplete.c 1.33-1.34 lib/libedit/readline.c 1.112-1.115 Man page improvements, fix overlapping strcpy, improve readline compatibility, clang build fix.
1.32.8.1	23-Jul-2017	snj	Pull up following revision(s) (requested by kre in ticket #102): lib/libedit/Makefile: 1.64-1.65 lib/libedit/editline.3: 1.94-1.96 lib/libedit/editrc.5: 1.33 lib/libedit/el.c: 1.93-1.94 lib/libedit/el.h: 1.42 lib/libedit/literal.c: 1.1-1.3 lib/libedit/literal.h: 1.1-1.2 lib/libedit/prompt.c: 1.27 lib/libedit/read.c: 1.103 lib/libedit/refresh.c: 1.52-1.54 lib/libedit/refresh.h: 1.11 lib/libedit/terminal.c: 1.33 Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming. -- Include EDITRC in doc. -- mention the limitation of the literal sequence delimiter. -- - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display. -- - add literal sequence handling. -- remove unused variable -- add literal escape sequence support, patterned after the tcsh ones. -- fix comment -- Fix an obvious, but almost invisible typo (avoid some core dumps). -- Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale). -- Remove workaround for ancient HTML generation code.
1.102	03-Jan-2025	rillig	libedit: remove redundant break statements after EL_ABORT
1.101	30-Oct-2022	christos	branches: 1.101.4; improvements in malloc/free handling.
1.100	15-Aug-2021	christos	Disable attempts to handle EINTR and non-blocking I/O by default. It is confusing to other programs and unexpected behavior. Reported by Ingo Schwarze. This behavior is now controlled with EL_SAFEREAD.
1.99	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.98	26-Apr-2019	christos	Follow the man page for EL_GETTC and not require a NULL terminated argument list: https://reviews.llvm.org/D61191
1.97	18-Nov-2018	christos	Remove utf-8 requirement (Yuichiro NAITO)
1.96	01-Jan-2018	christos	branches: 1.96.2; 1.96.4; Only FLUSH if we are ending libedit; DRAIN if we suspend for readline. This allows pasting multiline buffers (Gerry Swislow)
1.95	05-Sep-2017	christos	For readline emulation, don't reset the tty to "sane" (cooked) mode if we did not start this way. Also set and reset the tty on entry and exit from readline() since this is what readline does.
1.94	27-Jun-2017	christos	add literal escape sequence support, patterned after the tcsh ones.
1.93	27-Jun-2017	kre	Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming.
1.92	22-May-2016	christos	branches: 1.92.8; Stop the read module from poking the el_chared.c_macro data structure currently belonging to the chared module. The read module does so from three of its functions, while no other module uses the macro data, not even the chared module itself. That's quite logical because macros are a feature of input handling, all of which is done by the read module, and none by the chared module. So move the data into the read modules's own opaque data structure, struct el_read_t. That simplifies internal interfaces in several respects: The semi-public chared.h has one fewer struct, one fewer #define, and one fewer member in struct el_chared_t; all three move to one single C file, read.c, and are now module-local. And the internal interface function ch_reset() needs one fewer argument, making the code of many functions in various modules more readable. The price is one additional internal interface function, read_end(), 10 lines long including comments, called publicly from exactly one place: el_end() in el.c. That's hardly an increase in complexity since most other modules already have their *_end() function, read.c was the odd one out not having one. From Ingo Schwarze
1.91	09-May-2016	christos	s/protected/libedit_private/g
1.90	02-May-2016	christos	eliminate static buffer with custom resizing code.
1.89	19-Apr-2016	christos	From Ingo Schwarze: - Put the data type el_rfunc_t into the public header <histedit.h>. - Make el_read in struct editline an opaque pointer rather than an embedded struct. - Do not include "read.h" everywhere, but only in the two files needing access to el_read, read.c and el.c. - To functions that don't need more, pass the struct el_read_t * rather than the full EditLine *. - Of course, that means that read_init() can now fail from memory exhaustion, but it's easy to clean up after that.
1.88	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.87	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.86	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.85	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.84	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.83	24-Feb-2016	christos	Make the read_char function always take a wchar_t * argument (Ingo Schwarze)
1.82	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.81	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.80	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.79	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.78	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.77	15-Feb-2016	christos	Use getline for better portability.
1.76	15-Feb-2016	christos	use fparseln() to avoid needing to deal with missing \n in the last line and also to handle comments automatically.
1.75	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.74	08-Dec-2015	christos	Only reset the terminal if we have a tty (Boris Ranto)
1.73	18-Jun-2014	christos	Don't depend on weak aliases to define the vi "alias" expansion function, provide an API instead to set it.
1.72	22-Jan-2013	christos	branches: 1.72.6; provide an el_init_fd function.
1.71	11-Sep-2012	christos	branches: 1.71.2; PR/46942: Steffen Nurpmeso: editline(3): el_get(): fix UNBUFFERED return
1.70	11-Mar-2012	christos	use arraycount
1.69	18-Nov-2011	christos	remove unrecheable code (Kamil Dudka)
1.68	29-Jul-2011	christos	branches: 1.68.2; KNF return (\1); -> return \1;
1.67	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.66	28-Jul-2011	christos	Rename key to keymacro to avoid conflicts with term.h. The renaming of term to terminal was again to avoid conflicts with term.h. term.h is a moving namespace violation.
1.65	28-Jul-2011	christos	avoid shadowed variable
1.64	28-Jul-2011	christos	term -> terminal eliminate MAXPATHLEN
1.63	26-Jul-2011	christos	don't stop reading after empty lines from: Nirbhay Choubey
1.62	20-Mar-2011	bouyer	Fix use-after-free. Discovered by running tests with MALLOC_OPTIONS=J (should this be the default for tests ?), pointed out by njoly@
1.61	27-Jan-2011	christos	don't turn on editing if stdout is not a tty.
1.60	28-Aug-2010	christos	branches: 1.60.2; setup a callback to be invoked on resize buffers so that readline can reset rl_line_buffer which unfortunately some applications use it directly.
1.59	15-Apr-2010	christos	From Jess Thrysoee - use nl_langinfo to test for UTF-8, because some locales are UTF-8 without reflecting it in their names.
1.58	31-Dec-2009	christos	- Document and enable wide character support. - Fix read function compatibility.
1.57	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.56	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.55	25-Jul-2009	christos	Ignore comment lines in .editrc from Jess Thrysoee
1.54	22-Jul-2009	christos	Only need path if we have issetugid... From Anon Ymous
1.53	22-Jul-2009	christos	Move filename to the scope it is being used. From Michael Cook mcook at bbn dot com
1.52	11-May-2009	christos	restore binary compatibility by providing new prompt functions that take an extra literal character.
1.51	31-Mar-2009	christos	Implement literal prompt sequences. Now someone can implement RL_PROMPT_START_LITERAL/RL_PROMPT_END_LITERAL :-)
1.50	21-Feb-2009	christos	back out all prompt changes. they are not needed.
1.49	19-Feb-2009	cube	Order of evaluation of arguments is undefined, so call va_arg() in an explicit order. Fixes a segfault with bc reported by Patrick Welche on current-users.
1.48	17-Feb-2009	christos	allow for a prompt argument.
1.47	18-Jan-2009	lukem	branches: 1.47.2; fix -Wsign-compare issues
1.46	10-Sep-2008	christos	Allow a single process to control multiple ttys (for pthreads using _REENTRANT) using multiple EditLine objects. Mostly from Preston A. Elder.
1.45	05-Apr-2008	christos	branches: 1.45.4; add EL_REFRESH for the benefit of readline
1.44	15-Dec-2006	christos	add EL_GETFP, and EL_SETFP.
1.43	25-Nov-2006	freza	s/el->errfile/el->el_errfile/g in debug code, fixes MKDEBUGLIB build.
1.42	24-Nov-2006	christos	- Add more readline functions, enough for gdb-6.5 - Make el_get varyadic, and implement EL_GETTC. - XXX: the EL_SETTC api will change in the future.
1.41	19-Aug-2005	christos	PR/31012: Barry Naujok: libedit el_get with EL_EDITOR op does not work Fixed as suggested.
1.40	01-Aug-2005	christos	Don't reset the macro strings each time we enter el_gets(), otherwise el_push() is unusable programmatically.
1.39	08-Jul-2004	christos	PR/23486: Andreas Gustafsson: gdb no longer works with emacs - make sure that we keep previous contents of the buffer in unbuffered mode. - when turning editing on and off keep tty consistent.
1.38	27-Feb-2004	christos	branches: 1.38.2; Better fix for rl_prep_terminal() from Gerry Swislow.
1.37	21-Feb-2004	christos	Separate tty separation from the prompt printing function. From Gerry Swislow.
1.36	18-Oct-2003	christos	change allocation policy in el_push to allocate the string itself. fix issues with strdup.
1.35	16-Oct-2003	christos	More libedit readline emulation functions from: Gerry Swislow <gerry at certif dot com>
1.34	26-Sep-2003	christos	Implement enough of readline's 4.0 async mode to make gdb happy. This is not complete yet, but it seems to work... This required to introduce an unbuffered mode to el_gets(), but that was a minor change.
1.33	14-Sep-2003	christos	- provide enough hooks to compile gdb-5.3 - fix el_get(e, EL_TERMINAL, (char **))
1.32	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.31	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.30	12-Nov-2002	thorpej	Avoid strict alias warning.
1.29	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.28	03-Feb-2002	christos	Don't use HAVE_ yet.
1.27	03-Feb-2002	christos	Fix a warning.
1.26	08-Nov-2001	mycroft	Remove an unused variable.
1.25	02-Nov-2001	christos	If term_init() fails, cleanup and return NULL. This avoids other lossage. Pointed by charles.
1.24	09-Oct-2001	christos	PR/14188: Anthony Mallet: Provide an opaque data pointer to client programs.
1.23	27-Sep-2001	christos	PR/14067: Anthony Mallet: Provide a programmatic way to set the read_char function via a new el_set() operation. Thanks, nicely done :-)
1.22	24-Sep-2001	wiz	va_{start,end} audit: Make sure that each va_start has one and only one matching va_end, especially in error cases. If the va_list is used multiple times, do multiple va_starts/va_ends. If a function gets va_list as argument, don't let it use va_end (since it's the callers responsibility). Improved by comments from enami and christos -- thanks! Heimdal/krb4/KAME changes already fed back, rest to follow. Inspired by, but not not based on, OpenBSD.
1.21	05-Jan-2001	christos	depoison the pure editline code from readline compatibility hacks.
1.20	11-Nov-2000	christos	- add support for home and end keys. - improve debugging support
1.19	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.18	02-Aug-2000	chuck	make xxgdb and a gdb linked with libedit's readline emulation work properly together. xxgdb communicates with a gdb running on a pty that it sets to -echo,-onlcr prior to forking the gdb process. GNU readline preserves the -echo setting while libedit was undoing it (setting the tty to a sane state and totally confusing xxgdb's parser). this diff simply disables libedit if both readline emulation and "stty -echo" are used/set. that is enough to make xxgdb work once again, but (XXX) this is not how GNU readline handles stty -echo (it does not echo anything, but editing commands like ^A,^K, etc. still work), so the readline emulation isn't perfect. change reviewed by christos.
1.17	28-Jun-2000	sommerfeld	Only look in home directory for .editrc. (Discussed with Christos.)
1.16	15-May-2000	christos	branches: 1.16.4; don't dump core on empty files.
1.15	12-Nov-1999	lukem	- implement printing a right-side prompt. code derived from similar work I wrote for tcsh(1) three years ago. - implement EL_RPROMPT, which allows a setting/getting of a function which returns a string to be used as the right-side prompt. - improve HISTORY and AUTHORS sections in editline(3). - bump shlib minor version for EL_RPROMPT. XXX: due to an implementation issue, the rprompt has a 1 space gap before the edge of the logical screen. editline's logical screen is 1 space less than the full screen width, so there's a 2 space gap between the rprompt and the right end of the physical screen. i'm not concerned about this.
1.14	21-Sep-1999	lukem	branches: 1.14.4; * in el_source(): - don't reopen fp if it was successfully opened in a previous check - use strlcpy()/strlcat() instead of snprintf() to build a pathname, since the former are more portable to other systems than the latter * whitespace fascism
1.13	02-Jul-1999	simonb	More trailing white space.
1.12	12-Dec-1998	christos	branches: 1.12.2; delint
1.11	28-Sep-1998	christos	fix core-dump caused by maps not being initialized before tty is called.
1.10	29-Jul-1998	lukem	* add more checks for NULL pointers in passed arguments * implement el_get(EditLine , int op, void result), which does the inverse of el_set() * add EL_EDITMODE operation to el_set and el_get; if non zero editing is enabled (the default). * add "edit on \| off" editrc command, which modifies EL_EDITMODE. users can now add '*:edit off' in ~/.editrc as an advisory to disable editing. NOTE: at this time EL_EDITMODE is just an indication of the state of the 'edit' command. It's up to the application to check this after el_source() or el_parse() to determine if editing is still required.
1.9	20-May-1998	christos	el_init takes a third stream argument and add el_beep.
1.8	20-Dec-1997	christos	Small optimization. Don't call isatty() on every invocation of el_gets, but remember if the tty setup failed... Also trim the input line of trailing \r's.
1.7	06-Jul-1997	christos	Fix compiler warnings.
1.6	24-Apr-1997	christos	Handle properly the case where the last line in the sourced file does not have a trailing '\n'. From Jeffrey C Honig.
1.5	11-Apr-1997	christos	Portability fixes: __const -> const BADSIG -> SIG_ERR int flags -> u_int flags #if __STDC__ -> #ifdef __STDC__
1.4	20-Mar-1997	christos	Fix potential stack overflow; from Keith Bostic.
1.3	17-Jan-1997	lukem	fix el_source() - a block needed braces around it (the indenting fooled me)
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.12.2.1	29-Jun-2000	he	Pull up revision 1.17 (via patch, requested by sommerfeld): Only read .editrc from home directory.
1.14.4.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.16.4.2	02-Aug-2000	chuck	pull up 1.18 (approved by thorpej). allows a gdb linked with libedit's readline emulation to work properly with xxgdb.
1.16.4.1	28-Jun-2000	sommerfeld	Pull up 1.17 (approved by thorpej): only look for .editrc in /d1/sommerfeld
1.38.2.1	10-Jul-2004	tron	Pull up revision 1.39 (requested by christos in ticket #620): PR/23486: Andreas Gustafsson: gdb no longer works with emacs - make sure that we keep previous contents of the buffer in unbuffered mode. - when turning editing on and off keep tty consistent.
1.45.4.1	24-Sep-2008	wrstuden	Merge in changes between wrstuden-revivesa-base-2 and wrstuden-revivesa-base-3.
1.47.2.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.60.2.1	08-Feb-2011	bouyer	Sync with HEAD
1.68.2.3	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.68.2.2	30-Oct-2012	yamt	sync with head
1.68.2.1	17-Apr-2012	yamt	sync with head
1.71.2.2	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.71.2.1	25-Feb-2013	tls	resync with head
1.72.6.1	10-Aug-2014	tls	Rebase.
1.92.8.1	23-Jul-2017	snj	Pull up following revision(s) (requested by kre in ticket #102): lib/libedit/Makefile: 1.64-1.65 lib/libedit/editline.3: 1.94-1.96 lib/libedit/editrc.5: 1.33 lib/libedit/el.c: 1.93-1.94 lib/libedit/el.h: 1.42 lib/libedit/literal.c: 1.1-1.3 lib/libedit/literal.h: 1.1-1.2 lib/libedit/prompt.c: 1.27 lib/libedit/read.c: 1.103 lib/libedit/refresh.c: 1.52-1.54 lib/libedit/refresh.h: 1.11 lib/libedit/terminal.c: 1.33 Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming. -- Include EDITRC in doc. -- mention the limitation of the literal sequence delimiter. -- - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display. -- - add literal sequence handling. -- remove unused variable -- add literal escape sequence support, patterned after the tcsh ones. -- fix comment -- Fix an obvious, but almost invisible typo (avoid some core dumps). -- Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale). -- Remove workaround for ancient HTML generation code.
1.96.4.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.96.4.1	10-Jun-2019	christos	Sync with HEAD
1.96.2.1	26-Nov-2018	pgoyette	Sync with HEAD, resolve a couple of conflicts
1.101.4.1	02-Aug-2025	perseant	Sync with HEAD
1.48	03-Jan-2025	rillig	libedit: remove redundant break statements after EL_ABORT
1.47	17-May-2024	christos	branches: 1.47.2; When calling el_line make sure that we call the resizing function callback because el_line updates the legacy LineInfo structure and we need to notify that the cached copy of the the buffer has changed. Of course the resizing function can call el_line itself to update the buffer, so prevent recursion. Bug found by Peter Rufer at Arista.
1.46	15-Aug-2021	christos	Disable attempts to handle EINTR and non-blocking I/O by default. It is confusing to other programs and unexpected behavior. Reported by Ingo Schwarze. This behavior is now controlled with EL_SAFEREAD.
1.45	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.44	18-Nov-2018	christos	Remove utf-8 requirement (Yuichiro NAITO)
1.43	05-Sep-2017	christos	branches: 1.43.2; 1.43.4; For readline emulation, don't reset the tty to "sane" (cooked) mode if we did not start this way. Also set and reset the tty on entry and exit from readline() since this is what readline does.
1.42	27-Jun-2017	christos	add literal escape sequence support, patterned after the tcsh ones.
1.41	24-May-2016	christos	branches: 1.41.8; From Ingo Schwarze: Reduce obfuscation of errno handling. There is only one purpose non-local errno handling is needed for: Inside el_wgets(), several functions call down indirectly to el_wgetc(), many of them via the dispatch table. When el_wgetc() fails, it does properly report failure, but then various cleanup is done which may clobber errno. But when returning due to failure, el_wgets() wants to have errno set to the reason of the original read failure, not to the reason of some subsequent failure of some cleanup operation. So el_wgetc() needs to save errno, and if it's non-zero, el_wgets() needs to restore it on failure. This core logic is currently obscured by the fact that el_errno is set and inspected at some additional places where it isn't needed. Besides, since el_wgetc() and and el_wgets() are both in read.c, el_errno does not need to be in struct editline, it can and should be local to read.c in struct el_read_t. Let's look at what can be simplified. 1. keymacro_get() abuses el_errno instead of having a proper error return code. Adding that error return code is easy because node_trav() already detects the condition and an adequate code is already defined. Returning it, testing for it in read_getcmd(), and returning with error from there removes the need to inspect el_errno from el_wgets() after calling read_getcmd(). Note that resetting lastchar and cursor and clearing buffer[0] is irrelevant. The code returns from el_wgets() right afterwards. Outside el_wgets(), these variables are no longer relevant. When el_wgets() is called the next time, it will call ch_reset() anyway, resetting the two pointers. And as long as lastchar points to the beginning of the buffer, the contents of the buffer won't be used for anything. 2. read_getcmd() doesn't need to set el_errno again after el_wgetc() failure since el_wgetc() already did so. While here, remove the silly "if EOF or error" comments from the el_wgetc() return value tests. It's a public interface documented in a manual, so people working on the implementation can obviously be expected to know how it works. It's a case of count++; /* Increment count. / 3. In the two code paths of el_wgets() that lead up to "goto noedit", there is no need to save the errno because nothing that might change it happens before returning. For clarity, since el_wgets() is the function restoring the errno, also move initializing it to the same function. Finally, note that restoring errno when the saved value is zero is wrong. No library code is ever allowed to clear a previously set value of errno. Only application programs are allowed to do that, and even they usually don't need to do so, except when using certain ill-designed interfaces like strtol(3). I tested that the behaviour remains sane in the following cases, all during execution of el_wgets(3) and with a signal handler for USR1 installed without SA_RESTART. Enter some text and maybe move around a bit. Then send a USR1 signal. The signal gets processed, then read_char() resumes reading. Send another USR1 signal. Now el_wgets() sets errno=EINTR and returns -1. * Press Ctrl-V to activate ed-quoted-insert. Then send a USR1 signal. The signal gets processed, then read_char() resumes reading. Send another USR1 signal. ed_quoted_insert() returns ed_end_of_file(), i.e. CC_EOF, and el_wgets() returns 0. * Press a key starting a keyboard macro. Then send a USR1 signal. The signal gets processed, then read_char() resumes reading. Send another USR1 signal. Now el_wgets() sets errno=EINTR and returns -1. * Press : to enter builtin command mode. Start typing a command. Then send a USR1 signal. The signal gets processed, then read_char() resumes reading. Send another USR1 signal. Now c_gets() returns -1, ed_command() beeps and returns CC_REFRESH, and el_wgets() resumes operation as it should. I also tested with "el_set(el, EL_EDITMODE, 0)", and it returns the right value and sets errno correctly.
1.40	09-May-2016	christos	s/protected/libedit_private/g
1.39	02-May-2016	christos	eliminate static buffer with custom resizing code.
1.38	19-Apr-2016	christos	From Ingo Schwarze: - Put the data type el_rfunc_t into the public header <histedit.h>. - Make el_read in struct editline an opaque pointer rather than an embedded struct. - Do not include "read.h" everywhere, but only in the two files needing access to el_read, read.c and el.c. - To functions that don't need more, pass the struct el_read_t * rather than the full EditLine *. - Of course, that means that read_init() can now fail from memory exhaustion, but it's easy to clean up after that.
1.37	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.36	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.35	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.34	24-Feb-2016	christos	Make the read_char function always take a wchar_t * argument (Ingo Schwarze)
1.33	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.32	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.31	16-Feb-2016	christos	get rid of bool_t (Ingo Schwarze)
1.30	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.29	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.28	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.27	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.26	12-Feb-2016	christos	GC IGNORE_EXTCHARS and simplify code (Ingo Schwarze)
1.25	29-Jul-2011	christos	pass -Wconversion
1.24	28-Jul-2011	christos	Rename key to keymacro to avoid conflicts with term.h. The renaming of term to terminal was again to avoid conflicts with term.h. term.h is a moving namespace violation.
1.23	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.22	27-Jan-2011	christos	don't turn on editing if stdout is not a tty.
1.21	31-Dec-2009	christos	branches: 1.21.2; - Document and enable wide character support. - Fix read function compatibility.
1.20	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.19	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.18	15-Feb-2009	christos	don't restart on EINTR, instead return NULL immediately. From Anon Ymous
1.17	15-Dec-2006	christos	branches: 1.17.20; add EL_GETFP, and EL_SETFP.
1.16	18-Oct-2003	christos	change allocation policy in el_push to allocate the string itself. fix issues with strdup.
1.15	26-Sep-2003	christos	Implement enough of readline's 4.0 async mode to make gdb happy. This is not complete yet, but it seems to work... This required to introduce an unbuffered mode to el_gets(), but that was a minor change.
1.14	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.13	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.12	27-Oct-2002	christos	vi mode and memory fixes from david laight.
1.11	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.10	09-Oct-2001	christos	PR/14188: Anthony Mallet: Provide an opaque data pointer to client programs.
1.9	27-Sep-2001	christos	PR/14067: Anthony Mallet: Provide a programmatic way to set the read_char function via a new el_set() operation. Thanks, nicely done :-)
1.8	06-Jan-2001	jdolecek	el_line_t: make 'limit' const
1.7	11-Nov-2000	christos	- add support for home and end keys. - improve debugging support
1.6	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.5	12-Nov-1999	lukem	- implement printing a right-side prompt. code derived from similar work I wrote for tcsh(1) three years ago. - implement EL_RPROMPT, which allows a setting/getting of a function which returns a string to be used as the right-side prompt. - improve HISTORY and AUTHORS sections in editline(3). - bump shlib minor version for EL_RPROMPT. XXX: due to an implementation issue, the rprompt has a 1 space gap before the edge of the logical screen. editline's logical screen is 1 space less than the full screen width, so there's a 2 space gap between the rprompt and the right end of the physical screen. i'm not concerned about this.
1.4	29-Jul-1998	lukem	branches: 1.4.6; * add more checks for NULL pointers in passed arguments * implement el_get(EditLine , int op, void result), which does the inverse of el_set() * add EL_EDITMODE operation to el_set and el_get; if non zero editing is enabled (the default). * add "edit on \| off" editrc command, which modifies EL_EDITMODE. users can now add '*:edit off' in ~/.editrc as an advisory to disable editing. NOTE: at this time EL_EDITMODE is just an indication of the state of the 'edit' command. It's up to the application to check this after el_source() or el_parse() to determine if editing is still required.
1.3	20-Dec-1997	christos	Small optimization. Don't call isatty() on every invocation of el_gets, but remember if the tty setup failed... Also trim the input line of trailing \r's.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.4.6.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.17.20.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.21.2.1	08-Feb-2011	bouyer	Sync with HEAD
1.41.8.1	23-Jul-2017	snj	Pull up following revision(s) (requested by kre in ticket #102): lib/libedit/Makefile: 1.64-1.65 lib/libedit/editline.3: 1.94-1.96 lib/libedit/editrc.5: 1.33 lib/libedit/el.c: 1.93-1.94 lib/libedit/el.h: 1.42 lib/libedit/literal.c: 1.1-1.3 lib/libedit/literal.h: 1.1-1.2 lib/libedit/prompt.c: 1.27 lib/libedit/read.c: 1.103 lib/libedit/refresh.c: 1.52-1.54 lib/libedit/refresh.h: 1.11 lib/libedit/terminal.c: 1.33 Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming. -- Include EDITRC in doc. -- mention the limitation of the literal sequence delimiter. -- - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display. -- - add literal sequence handling. -- remove unused variable -- add literal escape sequence support, patterned after the tcsh ones. -- fix comment -- Fix an obvious, but almost invisible typo (avoid some core dumps). -- Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale). -- Remove workaround for ancient HTML generation code.
1.43.4.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.43.4.1	10-Jun-2019	christos	Sync with HEAD
1.43.2.1	26-Nov-2018	pgoyette	Sync with HEAD, resolve a couple of conflicts
1.47.2.1	02-Aug-2025	perseant	Sync with HEAD
1.38	17-May-2024	christos	When calling el_line make sure that we call the resizing function callback because el_line updates the legacy LineInfo structure and we need to notify that the cached copy of the the buffer has changed. Of course the resizing function can call el_line itself to update the buffer, so prevent recursion. Bug found by Peter Rufer at Arista.
1.37	11-Jan-2022	christos	PR/56618: Walter Lozano: Improve libedit compatibility with readline by implementing: rl_copy_text, rl_erase_empty_line, rl_message, rl_on_new_line, rl_replace_line, rl_restore_prompt, rl_save_prompt
1.36	15-Aug-2021	christos	Disable attempts to handle EINTR and non-blocking I/O by default. It is confusing to other programs and unexpected behavior. Reported by Ingo Schwarze. This behavior is now controlled with EL_SAFEREAD.
1.35	26-Apr-2019	christos	Follow the man page for EL_GETTC and not require a NULL terminated argument list: https://reviews.llvm.org/D61191
1.34	09-May-2016	christos	branches: 1.34.16; Elide gcc warning about intermediate const casts caused by visibility change.
1.33	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.32	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.31	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.30	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.29	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.28	28-Feb-2016	christos	remove 4 clause licenses.
1.27	24-Feb-2016	christos	Make the read_char function always take a wchar_t * argument (Ingo Schwarze)
1.26	24-Feb-2016	christos	Get split el_getc and el_wgetc completely and call el_wgetc internally. Change some character constants to they wide versions. (Ingo Schwarze)
1.25	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.24	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.23	15-Feb-2016	christos	OpenBSD eln.c rev. 1.3 2011/11/27 21:46:44 pascal kill a C++-style comment
1.22	14-Feb-2016	christos	From Ingo Schwarze: el_getc() for the WIDECHAR case, that is, the version in eln.c. For a UTF-8 locale, it is broken in four ways: 1. If the character read is outside the ASCII range, the function does an undefined cast from wchar_t to char. Even if wchar_t is internally represented as UCS-4, that is wrong and dangerous because characters beyond codepoint U+0255 get their high bits truncated, meaning that perfectly valid printable Unicode characters get mapped to arbitrary bytes, even the ASCII escape character for some Unicode characters. But wchar_t need not be implemented in terms of UCS-4, so the outcome of this function is undefined for any and all input. 2. If insufficient space is available for the result, the function fails to detect failure and returns garbage rather than -1 as specified in the documentation. 3. The documentation says that errno will be set on failure, but that doesn't happen either in the above case. 4. Even for ASCII characters, the results may be wrong if wchar_t is not using UCS-4.
1.21	12-Feb-2016	christos	Avoid c99 for now.
1.20	12-Feb-2016	christos	GC IGNORE_EXTCHARS and simplify code (Ingo Schwarze)
1.19	18-May-2015	christos	make el_gets() return the number of characters read in wide mode (not the number of wide characters) From khorben@ by FreeBSD: https://svnweb.freebsd.org/ports/head/devel/libedit/files/patch-src_eln.c?\ revision=382458&view=markup XXX: Pullup-7
1.18	24-Mar-2015	christos	save and restore IGNORE_EXTCHARS like we do in the getc case. From: https://bugzilla.redhat.com/attachment.cgi?id=1001894
1.17	18-Jun-2014	christos	branches: 1.17.2; Don't depend on weak aliases to define the vi "alias" expansion function, provide an API instead to set it.
1.16	20-May-2014	christos	Always NULL terminate the argv[] array. From OpenBSD.
1.15	26-Feb-2014	christos	branches: 1.15.2; Add missing EL_REFRESH
1.14	11-Mar-2012	christos	branches: 1.14.2; include the NULL in the argv conversion
1.13	16-Aug-2011	christos	branches: 1.13.2; re-enable -Wconversion
1.12	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.11	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.10	20-Jun-2011	mrg	various build fixes for gcc 4.5. from chuq. XXX i'm not sure all of these work properly wtf pointer aliasing, but there are no casts at least... the lib/libpuffs/puffs_priv.h is definately a real bug fix. from chuq.
1.9	04-Nov-2010	christos	branches: 1.9.4; PR/43998, PR/44021: In narrow history emulation, don't treat UTF-8 character sets specially as far as history goes since we always need to do the conversion from narrow [history] to wide [editline].
1.8	28-Aug-2010	christos	setup a callback to be invoked on resize buffers so that readline can reset rl_line_buffer which unfortunately some applications use it directly.
1.7	15-Apr-2010	christos	From Jess Thrysoee - NARROW_HISTORY and IGNORE_EXTCHARS should not take effect if locale is UTF-8 - account for multi byte char length in
1.6	20-Jan-2010	christos	PR/42646: Joachim Kuebart: Shell crashes in libedit when window size changes (SIGWINCH). Return NULL if el_gets() gets interrupted.
1.5	19-Jan-2010	christos	Fix wrapper for EL_EDITOR, from Michael L. Hitch
1.4	12-Jan-2010	christos	- call the mapping function directly instead of el_wset(). - save the strings passed to the mapping function so that they don't get re-used. This leaks. To fix it properly we could either pass a flag to free particular entries before re-using, or allocate all of them. Allocating all of them wastes memory, allocating some of them makes the code more complex. This fixes compatibility binding (shell tab completion for example)
1.3	31-Dec-2009	christos	- Document and enable wide character support. - Fix read function compatibility.
1.2	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.1	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.9.4.1	23-Jun-2011	cherry	Catchup with rmind-uvmplock merge.
1.13.2.2	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.13.2.1	17-Apr-2012	yamt	sync with head
1.14.2.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.15.2.1	10-Aug-2014	tls	Rebase.
1.17.2.1	13-May-2015	martin	Sync lib/libedit with head, requested by christos in #753: lib/libedit/Makefile 1.53 lib/libedit/chartype.h 1.13 lib/libedit/editline.3 1.83-1.84 lib/libedit/editrc.5 1.28-1.29 lib/libedit/eln.c 1.18 lib/libedit/filecomplete.c 1.33-1.34 lib/libedit/readline.c 1.112-1.115 Man page improvements, fix overlapping strcpy, improve readline compatibility, clang build fix.
1.34.16.1	10-Jun-2019	christos	Sync with HEAD
1.38	29-Jun-2024	christos	don't use oldc before it is set.
1.37	29-Jun-2024	christos	Retrieve the cursor position after calling c_insert, because c_insert could enlarge the line buffer making the old cursor position point to freed memory. From Robert Morris https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=279772
1.36	09-May-2016	christos	s/protected/libedit_private/g
1.35	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.34	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.33	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.32	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.31	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.30	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.29	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.28	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.27	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.26	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.25	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.24	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.23	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.22	15-Feb-2009	christos	pass lint on _LP64.
1.21	06-Mar-2006	christos	branches: 1.21.28; Print the actual eofc, instead of ^D\b\b. Change internal character decoding to prevent buffer oveflows.
1.20	08-Aug-2005	christos	Spelling mistakes and comment errors (from FreeBSD via Stefan Farfeleder; many thanks)
1.19	28-Oct-2004	dsl	Use (unsigned char) cast to sanitise arguments to ctype functions.
1.18	27-Oct-2004	dsl	Fix a load of international alphabet problems with isxxx() and toupper() Change isspace(char_ptr) to isspace(char_ptr & 0xff) so that the correct piece of memory is looked at for the bit mask. gcc optimises out the '& 0xff' (on i386 at least). Fixes problems found by gcc when the splurious (int) cast is removed from the #defines in ctype.h
1.17	13-Aug-2004	mycroft	Delete-previous-char and delete-next-char without an argument are not supposed to modify the yank buffer in Emacs. Make it so.
1.16	02-Nov-2003	christos	If the kill buffer is empty return normal. From Gerry Swislow gerry at certif dot com
1.15	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.14	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.13	31-Mar-2003	perry	em-gosmacs-traspose->em-gosmacs-transpose (Igor Sobrado, PR misc/19909)
1.12	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.11	27-Oct-2002	christos	vi mode and memory fixes from david laight.
1.10	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.9	10-Jan-2001	jdolecek	Enlarge editline buffers as needed to support arbitrary length lines. This also addresses lib/9712 by Phil Nelson.
1.8	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.7	02-Jul-1999	simonb	More trailing white space.
1.6	20-May-1998	christos	cast is*() arg to unsigned char
1.5	03-Feb-1998	perry	remove obsolete register declarations
1.4	06-Jul-1997	christos	Fix compiler warnings.
1.3	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.2	30-Aug-1994	cgd	fix for pr 420, from Christos.
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	branches: 1.1.1.1.2; libedit!
1.1.1.1.2.1	30-Aug-1994	cgd	from trunk; pr 420
1.21.28.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.73	25-Apr-2023	christos	pass lint.
1.72	03-Feb-2023	christos	Add a entry point for the complete function for FreeBSD compatibility with sh.
1.71	30-Oct-2022	christos	improvements in malloc/free handling.
1.70	12-Mar-2022	christos	Fix filename autocompletion for strings like a\)b An escaped character should unconditionally be skipped together with the character that does the escaping. For example, in "a\)b" only the ")b" part was skipped but then the loop stopped at the "\" since it's one of the characters listed in word_break. (Piotr P. Stefaniak)
1.69	26-Sep-2021	christos	- Completion should not add a quote at the end of the line to match an already quoted quote. (Piotr Stefaniak) - fix lint unconst warnings for strchr
1.68	05-May-2021	christos	PR/56147: Miroslav Lichvar: Avoid memory leak if strdup fails.
1.67	28-Mar-2021	christos	Only unescape when we are quoting and don't add a space if we are quoting (we already did) (Piotr Stefaniak)
1.66	28-Mar-2021	christos	Pass the unescaped filename the the append function so it has to do less work (for example it can call stat(2) directly (Piotr Stefaniak)
1.65	27-Mar-2021	christos	Add fn_complete2() that controls the quoting of the returned match. Before it was based on the heuristic that we were not supplied an attempted_completion_function, which worked well because programs that supplied that function were not shells and did not want/understand shell quoting. Recently though Piotr Stefaniak wanted to enhance command completion for the Bourne Shell and this could benefit quoting the returned command. This function adds an extra flags argument that controls that quoting.
1.64	05-Jan-2020	abhinav	PR lib/54510 - when user supplied completion function is there, don't unescape the string to be completed.
1.63	05-Jan-2020	tih	Summary: Remove over-simplified extraneous test The file name matching code in libedit tries to adjust to the presence of explicit " or ' characters in the input line, but tries too hard. Remove a conditional that goes overboard, and causes the completion code to fail if a quoted string is seen before the filename to be expanded, as in grep 'foo' bar<TAB> Before this change, the above would not expand any possible completions, even if they existed, because it would choose to look for files whose names started with " bar".
1.62	10-Dec-2019	christos	When 'attempted_completion_function' non-NULL, with a 'single_match' match, the expected space is not being added. Problem observed with "chronyc" and "sqlite3" tab completion. That functionality got moved to escape_filename() for the !attempted_completion_function case, but the non-NULL 'attempted_completion_function' case must also be handled. (Lonnie Abelbeck)
1.61	09-Oct-2019	christos	add +1 to strlcpy's (Patrick Welche)
1.60	08-Oct-2019	christos	remore error(1) comment
1.59	08-Oct-2019	christos	Change strncpy to either memcpy (when we know the len), or strlcpy (when we used to NUL terminate explicitly.
1.58	08-Sep-2019	abhinav	PR lib/54510: Fix file completion inside quotes which broke in rev 1.53 While there also fix handling character appending in the file completions when inside quotes. For example when inside a quote, if the completion is a directory then append a '/' but don't close the quote. On the other hand when inside a quote if the completion is a file name and it is the only match then we can close the quote.
1.57	28-Jul-2019	christos	PR/54415: Ricky Zhou: libedit stats completions for non-file completions Use the proper completion function and account for the character appended by the function when computing the number of columns.
1.56	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.55	20-Apr-2019	abhinav	PR lib/54131 - declare the loop variable outside the for loop
1.54	12-Apr-2019	christos	PR/54117: Paavo Helde: Fix memory overrun: Account for the closing quote in memory allocation if quoted.
1.53	31-Mar-2019	abhinav	Perform quoting of filename completions when there are multiple matches as well Quoting of special characters in filename completion was implemented for single match case, this enables it for multiple matches as well. For example: $ touch 'foo bar' $ touch 'foo baz' $ ls fo<TAB> autocompletes to => $ ls foo\ ba hitting <TAB> again shows: foo bar foo baz This required unescaping escape sequences generated during last completion in order to find the word to complete. While there, also update the test to include cases for multiple matches. Reviewed by christos
1.52	24-Mar-2019	abhinav	Only quote the completion matches if we are doing filename completion If the user supplies a value for the attempted_completion_function parameter then we cannot be sure if the completion is for filename or something else, in such a case don't attempt to quote the completion matches. Reviewed by christos This should address PR lib/54067
1.51	04-May-2018	christos	branches: 1.51.2; fix uninitialized
1.50	04-May-2018	abhinav	Handle filename autocompletion when the cursor is at a backslash or quote character For example, handle following case: $ touch 'foo bar' $ ls foo\<TAB> --> $ ls foo\ bar Also add test cases for this. Thanks to Christos for review
1.49	02-May-2018	abhinav	Add '*' and '[' to the list of characters which need escaping during autocompletion
1.48	27-Oct-2017	abhinav	branches: 1.48.2; Add NULL check after doing memory allocation at a couple of places ok christos@
1.47	15-Oct-2017	abhinav	Add support for escaping special characters when doing filename completion. For instance if the file name is "foo bar": $ ls foo<TAB> should get autocompleted to: $ ls foo\ bar Works for similar other characters too, which need escaping. Also, add an accompanying test program to ensure the escaping is correct in various scenarios (within quotes, without quotes, with other special characeters) Thanks to Christos for reviews, help and feedback.
1.46	16-Sep-2017	abhinav	Fix indentation (convert spaces to tab)
1.45	21-Apr-2017	abhinav	When doing filename autocompletion, append a trailing slash at the end of directory names. We already do this when there is only one completion option but in case of of multiple completion options, it wasn't being done. ok christos@
1.44	31-Oct-2016	abhinav	branches: 1.44.2; Fix file name auto completion in one specific case. For example if you do $mkdir -p /tmp/dir1/dir2 Then: $ls /tmp/di <TAB> auto completes to $ls /tmp/dir1/ Hitting <TAB> again auto completes to $ls /tmp/dir1/dir2 Whereas it should auto complete to $ls /tmp/dir1/dir2/ Essentially, in cases like above where you have to hit <TAB> twice to get to the match and there is only one match (because only one file/sub-directory) then auto complete doesn't work correctly. It doesn't append a trailing slash (in case of directory) or a space (in case of a file) to the match name. I have tested file name completion in sh(1) and symbol completion in gdb after this change.
1.43	11-Apr-2016	christos	branches: 1.43.2; Char -> wchar_t from Ingo Schwarze.
1.42	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.41	09-Apr-2016	christos	Change some 0's to NULL's from Pedro Giffuni
1.40	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.39	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.38	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.37	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.36	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.35	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.34	18-Oct-2014	riz	callers's -> caller's
1.33	18-Oct-2014	snj	src is too big these days to tolerate superfluous apostrophes. It's "its", people!
1.32	05-Jun-2014	christos	branches: 1.32.2; PR/48876: Dmitriy Grigoryev: Core dump in readline lib on attempted expansion Make sure we have 2 matches before calling strcmp().
1.31	16-Sep-2011	plunky	branches: 1.31.8; 1.31.18; NULL does not need a cast, here
1.30	16-Aug-2011	christos	re-enable -Wconversion
1.29	29-Jul-2011	christos	pass -Wconversion
1.28	29-Jul-2011	christos	- fix unused params - unconditionalize vis.h
1.27	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.26	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.25	28-Jul-2011	christos	fix unused variable warnings on systems without _r functions
1.24	28-Jul-2011	christos	eliminate alloca for portability portable getpw{nam,uid}
1.23	06-Dec-2010	dholland	Improve previous to avoid changing the interface of an externally exposed function. (But note that this function is neither documented nor declared in any installed header file, and it probably should not be externally exposed.) Related to PR 44183, closes PR 44186.
1.22	02-Dec-2010	dholland	add const, from PR 44183.
1.21	02-Dec-2010	dholland	Fix up bodgy code for printing completion matches; it used to sometimes skip entries, print (null), run off the end of the array, or occasionally receive SIGSEGV, and now will, hopefully at least, do none of that. Based in part on the patch in PR 44183 from Sergio Acereda; I also did some tidyup and fixed it to print top-to-bottom first like ls(1).
1.20	15-Nov-2010	christos	don't increment i twice in the loop. From Michael Byrnes
1.19	01-Jun-2010	christos	tidy up memory allocation and don't unnecessarily print "./" before names.
1.18	18-Jan-2010	christos	PR/42637: Joachim Kuebart: Shell tab completion crashes due to libedit stack smashing
1.17	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.16	28-Dec-2009	christos	improve on the listing display by printing only one character after the filename not two, and no trailing blanks. I will revisit this when I write the ls-F code.
1.15	16-Feb-2009	christos	fix sign compare issues.
1.14	15-Feb-2009	christos	pass lint on _LP64.
1.13	26-Jan-2009	apb	branches: 1.13.2; Define HAVE_STRUCT_DIRENT_D_NAMLEN in config,h, and test it when deciding whether to use entry->d_namlen or strlen(entry->d_name). Addresses PR 40477 by Robert Millan.
1.12	11-Jan-2009	christos	- insert a space after the recognized string if it was an exact match - initialize properly the string used for completion. From Alex Bligh alex at alex dot org dot uk - Make char constants consistent
1.11	29-Apr-2008	martin	Convert to new 2 clause license
1.10	09-Nov-2006	christos	branches: 1.10.16; don't use alloca with ssp.
1.9	21-Aug-2006	christos	Change to a 3 clause copyright after permission of the holders.
1.8	03-Aug-2005	christos	Fix reversed test; from Gerry Swislow
1.7	11-Jun-2005	christos	PR/30500: Paul Shupak: Inconsistent definition of tilde_expand(). Provide a layer of indirection between the readline compatibility functions and our internal implementation, so that we have the freedom to change the function signature.
1.6	10-Jun-2005	christos	Bug reported from Martin Dietze: The place to change the completion_append_character is usually somewhere in the `rl_completion_entry_function' callback which is where one usually can distinguish between file- or dir-like entries to append a slash for dirs etc. This does no longer work since `fn_complete()' takes the `append_character' as argument before the callback is executed, so that changes to the variable `rl_completion_append_character' have in fact no effect for the current completion. Fix by adding a function that returns the rl_completion_append_character, when it gets passed in a filename in readline emulation.
1.5	18-May-2005	christos	Make completion_matches non-static since readline wants it.
1.4	12-May-2005	christos	PR/30215: Kouichirou Hiratsuka: /bin/sh dumps core with tabcomplete Don't core-dump when trying to complete an empty line; instead assume ./
1.3	09-May-2005	dsl	Use getpwuid_r(getuid(), ...) to expand ~/.... Don't replace ~xyz with /home/xyz when expanding ~
1.2	07-May-2005	dsl	gdb directly calls filename_completion_function()
1.1	07-May-2005	dsl	Separate out the filename completion functions from the readline() code. Pass in loads of parameters instead of relying on shed-loads of global variables to modify the behaviour. The filename completion code can now be enabled by code that uses el_gets(). (eg /bin/sh)
1.10.16.1	18-May-2008	yamt	sync with head.
1.13.2.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.31.18.1	10-Aug-2014	tls	Rebase.
1.31.8.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.32.2.1	13-May-2015	martin	Sync lib/libedit with head, requested by christos in #753: lib/libedit/Makefile 1.53 lib/libedit/chartype.h 1.13 lib/libedit/editline.3 1.83-1.84 lib/libedit/editrc.5 1.28-1.29 lib/libedit/eln.c 1.18 lib/libedit/filecomplete.c 1.33-1.34 lib/libedit/readline.c 1.112-1.115 Man page improvements, fix overlapping strcpy, improve readline compatibility, clang build fix.
1.43.2.2	26-Apr-2017	pgoyette	Sync with HEAD
1.43.2.1	04-Nov-2016	pgoyette	Sync with HEAD
1.44.2.1	21-Apr-2017	bouyer	Sync with HEAD
1.48.2.1	21-May-2018	pgoyette	Sync with HEAD
1.51.2.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.51.2.1	10-Jun-2019	christos	Sync with HEAD
1.15	25-Apr-2023	christos	pass lint.
1.14	26-Sep-2021	christos	make flag unsigned to match prototype of the function used
1.13	28-Mar-2021	christos	document the flag
1.12	27-Mar-2021	christos	Add fn_complete2() that controls the quoting of the returned match. Before it was based on the heuristic that we were not supplied an attempted_completion_function, which worked well because programs that supplied that function were not shells and did not want/understand shell quoting. Recently though Piotr Stefaniak wanted to enhance command completion for the Bourne Shell and this could benefit quoting the returned command. This function adds an extra flags argument that controls that quoting.
1.11	21-Apr-2017	abhinav	When doing filename autocompletion, append a trailing slash at the end of directory names. We already do this when there is only one completion option but in case of of multiple completion options, it wasn't being done. ok christos@
1.10	11-Apr-2016	christos	branches: 1.10.2; 1.10.4; Char -> wchar_t from Ingo Schwarze.
1.9	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.8	16-Feb-2009	christos	fix sign compare issues.
1.7	15-Feb-2009	christos	pass lint on _LP64.
1.6	29-Apr-2008	martin	branches: 1.6.8; Convert to new 2 clause license
1.5	21-Aug-2006	christos	branches: 1.5.16; Change to a 3 clause copyright after permission of the holders.
1.4	11-Jun-2005	christos	PR/30500: Paul Shupak: Inconsistent definition of tilde_expand(). Provide a layer of indirection between the readline compatibility functions and our internal implementation, so that we have the freedom to change the function signature.
1.3	10-Jun-2005	christos	Bug reported from Martin Dietze: The place to change the completion_append_character is usually somewhere in the `rl_completion_entry_function' callback which is where one usually can distinguish between file- or dir-like entries to append a slash for dirs etc. This does no longer work since `fn_complete()' takes the `append_character' as argument before the callback is executed, so that changes to the variable `rl_completion_append_character' have in fact no effect for the current completion. Fix by adding a function that returns the rl_completion_append_character, when it gets passed in a filename in readline emulation.
1.2	07-May-2005	dsl	gdb directly calls filename_completion_function()
1.1	07-May-2005	dsl	Separate out the filename completion functions from the readline() code. Pass in loads of parameters instead of relying on shed-loads of global variables to modify the behaviour. The filename completion code can now be enabled by code that uses el_gets(). (eg /bin/sh)
1.5.16.1	18-May-2008	yamt	sync with head.
1.6.8.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.10.4.1	21-Apr-2017	bouyer	Sync with HEAD
1.10.2.1	26-Apr-2017	pgoyette	Sync with HEAD
1.34	23-Jul-2019	christos	put the NULL check immediately after the allocation
1.33	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.32	05-Mar-2017	christos	branches: 1.32.12; one extra char for NUL.
1.31	05-Mar-2017	christos	Grow the buffer for event search if there was not enough space. From Gerry Swislow
1.30	07-Nov-2016	christos	branches: 1.30.2; Change the way the built-in history works; some programs enter history with the trailing newline, others don't so don't make any assumptions about it when printing. Also print the correct event number (generated), separate the event number from the event with a tab, and visually encode the string (don't encode tabs and spaces though).
1.29	09-May-2016	christos	branches: 1.29.2; s/protected/libedit_private/g
1.28	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.27	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.26	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.25	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.24	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.23	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.22	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.21	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.20	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.19	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.18	28-Jul-2011	christos	fix confusion with wide functions.
1.17	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.16	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.15	01-Nov-2003	christos	Fixes from Otto Moerbeek otto at drijf dot net
1.14	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.13	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.12	21-Jan-2003	christos	Add a uniquefier for the history function.
1.11	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.10	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.9	17-May-2001	christos	PR/12963:Jason Waterman: Fix signed cast problems.
1.8	10-Jan-2001	jdolecek	Enlarge editline buffers as needed to support arbitrary length lines. This also addresses lib/9712 by Phil Nelson.
1.7	04-Jan-2001	christos	consistently check for allocation failures and return -1, if we could not get more memory.
1.6	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.5	02-Jul-1999	simonb	More trailing white space.
1.4	14-Oct-1997	christos	PR/4257: Jaromir Dolecek: history() has no generic error handling and isn't reentrant. This changes the interface of the history function, so we need a major number bump.
1.3	06-Jul-1997	christos	Fix compiler warnings.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.29.2.2	20-Mar-2017	pgoyette	Sync with HEAD
1.29.2.1	07-Jan-2017	pgoyette	Sync with HEAD. (Note that most of these changes are simply $NetBSD$ tag issues.)
1.30.2.1	21-Apr-2017	bouyer	Sync with HEAD
1.32.12.1	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.23	01-Sep-2017	christos	PR/51517: Jay West: Tty settings not restored on exit PR/51518: Jay West: prompt is interleaved with client output Both these issues are caused by rl_restore_handler not DTRT; fix it so that it kills the internal libedit state completely. This is inefficient, but it works. Also fix: 1. add append_history()/H_NSAVE_FP 2. call the rl_startup_hook before printing the first prompt as documented. callint it from rl_initialize breaks python, because the callback ends up being invoked before the readline module is installed, and we end up dereferencing a NULL pointer. 3. add el_resize_terminal. With those changes, s/lreadline/ledit/g in python works.
1.22	09-May-2016	christos	s/protected/libedit_private/g
1.21	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.20	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.19	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.18	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.17	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.16	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.15	30-Jan-2016	christos	Whitespace fix (Ingo Schwarze)
1.14	11-May-2014	christos	Add a history function that takes a FILE pointer; needed for Capsicum. From Eitan Adler
1.13	28-Jul-2011	christos	branches: 1.13.2; 1.13.8; 1.13.18; kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.12	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.11	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.10	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.9	21-Jan-2003	christos	Add a uniquefier for the history function.
1.8	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.7	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.6	10-Jan-2001	jdolecek	Enlarge editline buffers as needed to support arbitrary length lines. This also addresses lib/9712 by Phil Nelson.
1.5	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.4	14-Oct-1997	christos	PR/4257: Jaromir Dolecek: history() has no generic error handling and isn't reentrant. This changes the interface of the history function, so we need a major number bump.
1.3	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.2	15-Oct-1995	christos	Added history load and save to file functions.
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.13.18.1	10-Aug-2014	tls	Rebase.
1.13.8.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.13.2.1	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.62	03-Feb-2023	christos	Add a entry point for the complete function for FreeBSD compatibility with sh.
1.61	08-Feb-2022	rillig	libedit: fix typos, apply KNF to newly imported code (PR/56693) No binary change.
1.60	08-Feb-2022	christos	PR/56693: Walter Lozano: Add support for rl_delete_text and rl_set_key
1.59	11-Jan-2022	christos	PR/56618: Walter Lozano: Improve libedit compatibility with readline by implementing: rl_copy_text, rl_erase_empty_line, rl_message, rl_on_new_line, rl_replace_line, rl_restore_prompt, rl_save_prompt
1.58	15-Aug-2021	christos	Disable attempts to handle EINTR and non-blocking I/O by default. It is confusing to other programs and unexpected behavior. Reported by Ingo Schwarze. This behavior is now controlled with EL_SAFEREAD.
1.57	01-Sep-2017	christos	PR/51517: Jay West: Tty settings not restored on exit PR/51518: Jay West: prompt is interleaved with client output Both these issues are caused by rl_restore_handler not DTRT; fix it so that it kills the internal libedit state completely. This is inefficient, but it works. Also fix: 1. add append_history()/H_NSAVE_FP 2. call the rl_startup_hook before printing the first prompt as documented. callint it from rl_initialize breaks python, because the callback ends up being invoked before the readline module is installed, and we end up dereferencing a NULL pointer. 3. add el_resize_terminal. With those changes, s/lreadline/ledit/g in python works.
1.56	19-Apr-2016	christos	From Ingo Schwarze: - Put the data type el_rfunc_t into the public header <histedit.h>. - Make el_read in struct editline an opaque pointer rather than an embedded struct. - Do not include "read.h" everywhere, but only in the two files needing access to el_read, read.c and el.c. - To functions that don't need more, pass the struct el_read_t * rather than the full EditLine *. - Of course, that means that read_init() can now fail from memory exhaustion, but it's easy to clean up after that.
1.55	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.54	16-Feb-2016	christos	- don't set _GNU_SOURCE. We are not supposed to make decisions for others. - don't special-case wcsdup() From Ingo Schwarze.
1.53	18-Jun-2014	christos	Don't depend on weak aliases to define the vi "alias" expansion function, provide an API instead to set it.
1.52	11-May-2014	christos	Add a history function that takes a FILE pointer; needed for Capsicum. From Eitan Adler
1.51	12-Jul-2013	christos	branches: 1.51.4; Add a function to move the cursor.
1.50	22-Jan-2013	christos	provide an el_init_fd function.
1.49	31-May-2012	christos	branches: 1.49.2; remove stdint.h; it is not used.
1.48	28-Jul-2011	christos	branches: 1.48.2; kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.47	28-Aug-2010	christos	setup a callback to be invoked on resize buffers so that readline can reset rl_line_buffer which unfortunately some applications use it directly.
1.46	15-Apr-2010	christos	From Jess Thrysoee: add ifndef around def of _GNU_SOURCE
1.45	03-Jan-2010	christos	rename historyw -> history_w for consistency. add wide tst code and make it the default.
1.44	31-Dec-2009	christos	- Document and enable wide character support. - Fix read function compatibility.
1.43	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.42	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.41	07-Sep-2009	christos	apply apple patches from: http://opensource.apple.com/source/libedit/libedit-11/patches/
1.40	11-May-2009	christos	restore binary compatibility by providing new prompt functions that take an extra literal character.
1.39	31-Mar-2009	christos	Implement literal prompt sequences. Now someone can implement RL_PROMPT_START_LITERAL/RL_PROMPT_END_LITERAL :-)
1.38	21-Feb-2009	christos	back out all prompt changes. they are not needed.
1.37	19-Feb-2009	christos	bump version for prompt arg.
1.36	17-Feb-2009	christos	allow for a prompt argument.
1.35	05-Feb-2009	christos	branches: 1.35.2; match documentation in el_push
1.34	05-Apr-2008	christos	bump minor.
1.33	05-Apr-2008	christos	add EL_REFRESH for the benefit of readline
1.32	10-Jun-2007	christos	Fix tab/space confusion; from Stefan Farfeleder
1.31	15-Dec-2006	christos	add EL_GETFP, and EL_SETFP.
1.30	24-Nov-2006	christos	bump minor.
1.29	24-Nov-2006	christos	- Add more readline functions, enough for gdb-6.5 - Make el_get varyadic, and implement EL_GETTC. - XXX: the EL_SETTC api will change in the future.
1.28	14-Jul-2005	christos	PR/30747: David N. Williams: libedit is missing remove_history() Added, please test.
1.27	12-Jun-2005	christos	make this useable from c++.
1.26	07-May-2005	dsl	Separate out the filename completion functions from the readline() code. Pass in loads of parameters instead of relying on shed-loads of global variables to modify the behaviour. The filename completion code can now be enabled by code that uses el_gets(). (eg /bin/sh)
1.25	05-Dec-2003	lukem	Tokenization function enhancements: * Make tok_init(), tok_end(), tok_reset(), tok_line() and tok_str() publically available in <histedit.h> * Documented the public functions in editline(3) * Renamed tok_line() -> tok_str() * Added new tok_line() which takes a "const LineInfo " instead of "const char " (the former has "cursor" information), and optionally return the argv index ("int cursorc") and offset within that index ("int cursorv"). This means that completion routines can use the tokenization code to crack the line and easily find which word the cursor is at. (mmm, context sensitive completion :) * Fixed TEST/test.c when using "continuation" lines (unmatched quote or \ at EOL), and added some more DEBUG messages including highlighting where the cursor is (with a `_').
1.24	16-Oct-2003	christos	More libedit readline emulation functions from: Gerry Swislow <gerry at certif dot com>
1.23	26-Sep-2003	christos	Implement enough of readline's 4.0 async mode to make gdb happy. This is not complete yet, but it seems to work... This required to introduce an unbuffered mode to el_gets(), but that was a minor change.
1.22	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.21	21-Jan-2003	christos	Add a uniquefier for the history function.
1.20	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.19	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.18	09-Oct-2001	christos	PR/14188: Anthony Mallet: Provide an opaque data pointer to client programs.
1.17	27-Sep-2001	christos	PR/14067: Anthony Mallet: Provide a programmatic way to set the read_char function via a new el_set() operation. Thanks, nicely done :-)
1.16	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.15	28-Feb-2000	chopps	el_insertstr takes a "const char " not "char " now as it doesn't modify the argument.
1.14	12-Nov-1999	lukem	- implement printing a right-side prompt. code derived from similar work I wrote for tcsh(1) three years ago. - implement EL_RPROMPT, which allows a setting/getting of a function which returns a string to be used as the right-side prompt. - improve HISTORY and AUTHORS sections in editline(3). - bump shlib minor version for EL_RPROMPT. XXX: due to an implementation issue, the rprompt has a 1 space gap before the edge of the logical screen. editline's logical screen is 1 space less than the full screen width, so there's a 2 space gap between the rprompt and the right end of the physical screen. i'm not concerned about this.
1.13	20-Sep-1999	lukem	branches: 1.13.4; more whitespace fascism
1.12	02-Jul-1999	simonb	More trailing white space.
1.11	20-Dec-1998	kleink	Change multiple inclusion protection symbol name to NetBSD convention which doesn't violate namespace rules.
1.10	29-Jul-1998	lukem	* add more checks for NULL pointers in passed arguments * implement el_get(EditLine , int op, void result), which does the inverse of el_set() * add EL_EDITMODE operation to el_set and el_get; if non zero editing is enabled (the default). * add "edit on \| off" editrc command, which modifies EL_EDITMODE. users can now add '*:edit off' in ~/.editrc as an advisory to disable editing. NOTE: at this time EL_EDITMODE is just an indication of the state of the 'edit' command. It's up to the application to check this after el_source() or el_parse() to determine if editing is still required.
1.9	01-Jun-1998	lukem	* implement CC_REFRESH_BEEP; as per CC_REFRESH but beep as well. this is useful in completion when a partial completion is found * remove entry in BUGS about el_parse(); that was fixed a while ago
1.8	20-May-1998	christos	Add H_APPEND to simplify the interface.
1.7	20-May-1998	christos	- add extra argument to el_init - fix history functions - add el_beep()
1.6	14-Oct-1997	christos	PR/4257: Jaromir Dolecek: history() has no generic error handling and isn't reentrant. This changes the interface of the history function, so we need a major number bump.
1.5	11-Apr-1997	christos	Portability fixes: __const -> const BADSIG -> SIG_ERR int flags -> u_int flags #if __STDC__ -> #ifdef __STDC__
1.4	14-Jan-1997	lukem	Implement CC_REDISPLAY, which (unlink CC_REFRESH) redraws the entire input line (a la ^R). This is useful if the binding outputs information and mucks up the input line. To be used in ``list-choices'' bindings (refer to the ^D binding in csh when filec is set)
1.3	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.2	15-Oct-1995	christos	Added history load and save to file functions.
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.13.4.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.35.2.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.48.2.2	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.48.2.1	30-Oct-2012	yamt	sync with head
1.49.2.2	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.49.2.1	25-Feb-2013	tls	resync with head
1.51.4.1	10-Aug-2014	tls	Rebase.
1.64	11-Jul-2024	kre	Don't fchmod(fileno(fp), ...) in history_save_fp(). There are two reasons for this, first, the permissions of the history file should be able to be set by the user, not forced to 0600 every time the history file is overwritten (or appended to). And more importantly, the fp used for fileno(fp) might have come from fmemopen() or funopen() (etc) - none of which put a file descriptor in the "fd" field (ie: fileno(fp) == -1). To compensate for that, when a history file is opened (in history_save()) set the default permissions then - if the file is actually created. As fopen() cannot do that (it simply uses 0666&~umask) create the (normal type) of fp using (approximately) fdopen(open(...), ...) where the open supplies the 0600 default permissions that are desired here (which might still be restricted even more by the umask). Callers using history(...,H_SAVE_FP,...) or history(...,H_NSAVE_FP,...) now need to look after any permission setting required themselves (but as the doc says absolutely nothing about that, one way or the other, what happens in this area has always been unspecified, and still is) One "feature" of the fchmod() method is lost here - apart from forcing the 0600 permissions (which isn't really desirable) that fchmod() would also have failed if the current (effective) uid is not the owner of the history file (or root). If that is required, a test for it could be added later - the effect would be as it has always been, the file named must have been writable (or its directory writable if the file did not exist) the open would occur (potentially truncating the file) after which the fchmod() would be attempted, possibly failing, and if so, never writing anything. Any new uid test would work the same way. OK christos@
1.63	08-Oct-2019	christos	branches: 1.63.10; Change strncpy to either memcpy (when we know the len), or strlcpy (when we used to NUL terminate explicitly.
1.62	13-Sep-2018	kre	Fix editing mistake, remove ) from func call that is now gone.
1.61	13-Sep-2018	christos	more efficient to use decode_result :-) Pointed out by kre@
1.60	12-Sep-2018	christos	PR/53597: Yasuhiro Horimoto: Avoid segmentation fault in bad history file.
1.59	23-Dec-2017	uwe	branches: 1.59.2; 1.59.4; The order in which the arguments to a function are evaluated is undefined, so don't use va_arg() twice. PR lib/52849
1.58	01-Sep-2017	christos	PR/51517: Jay West: Tty settings not restored on exit PR/51518: Jay West: prompt is interleaved with client output Both these issues are caused by rl_restore_handler not DTRT; fix it so that it kills the internal libedit state completely. This is inefficient, but it works. Also fix: 1. add append_history()/H_NSAVE_FP 2. call the rl_startup_hook before printing the first prompt as documented. callint it from rl_initialize breaks python, because the callback ends up being invoked before the readline module is installed, and we end up dereferencing a NULL pointer. 3. add el_resize_terminal. With those changes, s/lreadline/ledit/g in python works.
1.57	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.56	11-Apr-2016	christos	chartype cleanups from Ingo Schwarze: - The file tokenizer.c no longer uses chartype.h, so don't include the header. - The dummy definitions of ct_{de,en}code_string() for the NARROWCHAR case are only used in history.c, so move them there. - Now the whole content of chartype.h is for the wide character case only. So remove the NARROWCHAR ifdef and include the header only in the wide character case. - In chartype.h, move ct_encode_char() below the comment explaining it. - No more need for underscores before ct_{de,en}code_string(). - Make the conversion buffer resize functions private. They are only called from the decoding and encoding functions inside chartype.c, and no need can possibly arise to call them from anywhere else.
1.55	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.54	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.53	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.52	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.51	15-Feb-2016	christos	Don't free getline memory (Ingo Schwarze).
1.50	15-Feb-2016	christos	Use getline for better portability.
1.49	15-Feb-2016	christos	Change the test for the size of encoded buffer to include the NULL, from OpenBSD; no functional change.
1.48	15-Feb-2016	christos	Use fparseln to avoid newline hacks.
1.47	11-May-2014	christos	Add a history function that takes a FILE pointer; needed for Capsicum. From Eitan Adler
1.46	18-Nov-2011	christos	branches: 1.46.6; 1.46.16; Add coverity annotations about unreachable code (Kamil Dudka)
1.45	29-Jul-2011	christos	branches: 1.45.2; pass -Wconversion
1.44	29-Jul-2011	christos	- fix unused params - unconditionalize vis.h
1.43	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.42	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.41	28-Jul-2011	christos	include vis.h from dot.
1.40	27-Jul-2011	christos	fix narrow compilation
1.39	27-Jul-2011	christos	- don't leave cursor dangling on memory failure or after clearing the list - compute the string length to be strvis'ed after the string is encoded
1.38	16-Jan-2011	christos	off by one in fetching history data. From: Gerry Swislow
1.37	03-Jan-2010	christos	rename historyw -> history_w for consistency. add wide tst code and make it the default.
1.36	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.35	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.34	07-Sep-2009	christos	apply apple patches from: http://opensource.apple.com/source/libedit/libedit-11/patches/
1.33	06-Feb-2009	sketch	Plug memory leak, from MySQL.
1.32	28-Sep-2006	christos	Fix memory leak found by valgrind (Julien Torres)
1.31	01-Aug-2005	christos	Don't forget to initialize h_del; from Julien Torres.
1.30	14-Jul-2005	christos	PR/30747: David N. Williams: libedit is missing remove_history() Added, please test.
1.29	06-Jul-2005	christos	Fix memory leaks found by valgrind. From Julien Torres
1.28	27-Nov-2004	christos	PR/26785: Jess Thrysoee: libedit - H_NEXT and H_PREV shifts cursor on failure
1.27	24-Aug-2004	christos	make sure that we round up to 1K.
1.26	20-Aug-2004	christos	PR/26725: Sergey S. Kostyliov: Typo in libedit, possible buffer overflow in src/lib/libedit/history.c:history_save()
1.25	18-Oct-2003	christos	change allocation policy in el_push to allocate the string itself. fix issues with strdup.
1.24	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.23	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.22	21-Jan-2003	christos	Add a uniquefier for the history function.
1.21	27-Oct-2002	christos	don't crash in memory shortage conditions.
1.20	13-Oct-2002	christos	write the vis(3) converted string into the file, not the original one. Noted by Tim Robbins. Hi luke!
1.19	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.18	29-Sep-2001	jdolecek	history_def_enter: fix off-by-one mistake in delete condition (the behaviour to keep at least one entry on the history list is retained) This fixes lib/9704 by Phil Nelson.
1.17	20-Mar-2001	christos	chmod the history file to 600 so that only the owner can read it. [inspired by the openbsd fix to readline]
1.16	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.15	29-Aug-2000	lukem	history_def_set has a `const int' as a third arg, not an `int'. picked up by the ultrix compiler, reported by simonb@ ...
1.14	02-Jul-1999	simonb	More trailing white space.
1.13	06-Mar-1999	mycroft	Add missing , in he_errlist[] initializer. Yay lint.
1.12	05-Feb-1999	christos	Encode the history lines using strvis() before saving it. Fixes gdb history problem with missing newlines.
1.11	12-Dec-1998	christos	- missing error message from array could cause core dump - delint
1.10	03-Dec-1998	cgd	per e-mail with christos: history_next_event() should return -1, not NULL, in case of error, just like history_prev_event() does.
1.9	20-May-1998	christos	Add H_APPEND to simplify the interface.
1.8	20-May-1998	christos	Add a function to be able to set the cursor to a given event number.
1.7	14-Oct-1997	christos	PR/4257: Jaromir Dolecek: history() has no generic error handling and isn't reentrant. This changes the interface of the history function, so we need a major number bump.
1.6	06-Jul-1997	christos	Fix compiler warnings.
1.5	11-Apr-1997	christos	Portability fixes: __const -> const BADSIG -> SIG_ERR int flags -> u_int flags #if __STDC__ -> #ifdef __STDC__
1.4	23-Jan-1997	mrg	- convert unsafe strcpy(), strcat() and sprintf() to the `n' versions. - some KNF.
1.3	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.2	15-Oct-1995	christos	Added history load and save to file functions.
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.45.2.2	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.45.2.1	17-Apr-2012	yamt	sync with head
1.46.16.1	10-Aug-2014	tls	Rebase.
1.46.6.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.59.4.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.59.4.1	10-Jun-2019	christos	Sync with HEAD
1.59.2.1	30-Sep-2018	pgoyette	Ssync with HEAD
1.63.10.1	02-Aug-2025	perseant	Sync with HEAD
1.1	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.24	28-Jul-2011	christos	Rename key to keymacro to avoid conflicts with term.h. The renaming of term to terminal was again to avoid conflicts with term.h. term.h is a moving namespace violation.
1.23	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.22	21-Feb-2009	christos	more size_t stuff.
1.21	16-Feb-2009	christos	fix sign compare issues.
1.20	15-Feb-2009	christos	pass lint on _LP64.
1.19	23-Mar-2006	christos	branches: 1.19.28; move declaration to header file.
1.18	06-Mar-2006	christos	Print the actual eofc, instead of ^D\b\b. Change internal character decoding to prevent buffer oveflows.
1.17	08-Aug-2005	christos	Spelling mistakes and comment errors (from FreeBSD via Stefan Farfeleder; many thanks)
1.16	06-Jul-2005	christos	Fix memory leaks found by valgrind. From Julien Torres
1.15	18-Oct-2003	christos	change allocation policy in el_push to allocate the string itself. fix issues with strdup.
1.14	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.13	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.12	17-May-2001	christos	PR/12963:Jason Waterman: Fix signed cast problems.
1.11	23-Jan-2001	jdolecek	sprinkle couple const
1.10	04-Jan-2001	christos	consistently check for allocation failures and return -1, if we could not get more memory.
1.9	11-Nov-2000	christos	- add support for home and end keys. - improve debugging support
1.8	04-Oct-2000	sommerfeld	format string audit (silence warnings, save space)
1.7	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.6	19-Jan-2000	christos	PR/9244: Kevin Schoedel: libedit dumps bindings inconsistently
1.5	02-Jul-1999	simonb	More trailing white space.
1.4	12-Dec-1998	christos	delint
1.3	06-Jul-1997	christos	Fix compiler warnings.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.19.28.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.14	28-Jul-2011	christos	Rename key to keymacro to avoid conflicts with term.h. The renaming of term to terminal was again to avoid conflicts with term.h. term.h is a moving namespace violation.
1.13	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.12	21-Feb-2009	christos	more size_t stuff.
1.11	15-Feb-2009	christos	pass lint on _LP64.
1.10	23-Mar-2006	christos	branches: 1.10.28; move declaration to header file.
1.9	06-Mar-2006	christos	Print the actual eofc, instead of ^D\b\b. Change internal character decoding to prevent buffer oveflows.
1.8	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.7	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.6	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.5	23-Jan-2001	jdolecek	sprinkle couple const
1.4	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.3	02-Jul-1999	simonb	More trailing white space.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.10.28.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.25	03-Jan-2025	rillig	libedit: remove redundant break statements after EL_ABORT
1.24	23-Jul-2019	christos	branches: 1.24.12; PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.23	24-May-2016	christos	branches: 1.23.16; From Ingo Schwarze: Reduce obfuscation of errno handling. There is only one purpose non-local errno handling is needed for: Inside el_wgets(), several functions call down indirectly to el_wgetc(), many of them via the dispatch table. When el_wgetc() fails, it does properly report failure, but then various cleanup is done which may clobber errno. But when returning due to failure, el_wgets() wants to have errno set to the reason of the original read failure, not to the reason of some subsequent failure of some cleanup operation. So el_wgetc() needs to save errno, and if it's non-zero, el_wgets() needs to restore it on failure. This core logic is currently obscured by the fact that el_errno is set and inspected at some additional places where it isn't needed. Besides, since el_wgetc() and and el_wgets() are both in read.c, el_errno does not need to be in struct editline, it can and should be local to read.c in struct el_read_t. Let's look at what can be simplified. 1. keymacro_get() abuses el_errno instead of having a proper error return code. Adding that error return code is easy because node_trav() already detects the condition and an adequate code is already defined. Returning it, testing for it in read_getcmd(), and returning with error from there removes the need to inspect el_errno from el_wgets() after calling read_getcmd(). Note that resetting lastchar and cursor and clearing buffer[0] is irrelevant. The code returns from el_wgets() right afterwards. Outside el_wgets(), these variables are no longer relevant. When el_wgets() is called the next time, it will call ch_reset() anyway, resetting the two pointers. And as long as lastchar points to the beginning of the buffer, the contents of the buffer won't be used for anything. 2. read_getcmd() doesn't need to set el_errno again after el_wgetc() failure since el_wgetc() already did so. While here, remove the silly "if EOF or error" comments from the el_wgetc() return value tests. It's a public interface documented in a manual, so people working on the implementation can obviously be expected to know how it works. It's a case of count++; /* Increment count. / 3. In the two code paths of el_wgets() that lead up to "goto noedit", there is no need to save the errno because nothing that might change it happens before returning. For clarity, since el_wgets() is the function restoring the errno, also move initializing it to the same function. Finally, note that restoring errno when the saved value is zero is wrong. No library code is ever allowed to clear a previously set value of errno. Only application programs are allowed to do that, and even they usually don't need to do so, except when using certain ill-designed interfaces like strtol(3). I tested that the behaviour remains sane in the following cases, all during execution of el_wgets(3) and with a signal handler for USR1 installed without SA_RESTART. Enter some text and maybe move around a bit. Then send a USR1 signal. The signal gets processed, then read_char() resumes reading. Send another USR1 signal. Now el_wgets() sets errno=EINTR and returns -1. * Press Ctrl-V to activate ed-quoted-insert. Then send a USR1 signal. The signal gets processed, then read_char() resumes reading. Send another USR1 signal. ed_quoted_insert() returns ed_end_of_file(), i.e. CC_EOF, and el_wgets() returns 0. * Press a key starting a keyboard macro. Then send a USR1 signal. The signal gets processed, then read_char() resumes reading. Send another USR1 signal. Now el_wgets() sets errno=EINTR and returns -1. * Press : to enter builtin command mode. Start typing a command. Then send a USR1 signal. The signal gets processed, then read_char() resumes reading. Send another USR1 signal. Now c_gets() returns -1, ed_command() beeps and returns CC_REFRESH, and el_wgets() resumes operation as it should. I also tested with "el_set(el, EL_EDITMODE, 0)", and it returns the right value and sets errno correctly.
1.22	09-May-2016	christos	s/protected/libedit_private/g
1.21	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.20	12-Apr-2016	christos	From Ingo Schwarze: * Delete the stubs of the XK_EXE mechanism that was never implemented. From a security, stability, and simplicity perspective, i would consider implementing it a truly terrible idea, so let's better get rid of it. * Do not use the local variable "num" in el_wgets() alternately for two completely different purposes. Only use it for the number of characters read, as stated in the comment (or -1 as long as that number is still unknown), not for the (more or less boolean) return value of read_getcmd(). Actually, there is no need at all to save the latter return value after testing it once. * The function read_getcmd() has very unusual return values: It returns -1 for success and 0 for EOF/error. Switch that around to 0 for success and -1 for EOF/error to be less confusing, and get rid of the OKCMD preprocessor macro. * Get rid of one #ifdef section in el_wgets() by using el->el_chared.c_macro directly at the only place where it is used. * Delete the unused MIN() macro.
1.19	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.18	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.17	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.16	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.15	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.14	24-Feb-2016	christos	Get split el_getc and el_wgetc completely and call el_wgetc internally. Change some character constants to they wide versions. (Ingo Schwarze)
1.13	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.12	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.11	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.10	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.9	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.8	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.7	16-Aug-2011	christos	re-enable -Wconversion
1.6	29-Jul-2011	christos	pass -Wconversion
1.5	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.4	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.3	28-Jul-2011	christos	whitespace
1.2	28-Jul-2011	christos	whitespace
1.1	28-Jul-2011	christos	Rename key to keymacro to avoid conflicts with term.h. The renaming of term to terminal was again to avoid conflicts with term.h. term.h is a moving namespace violation.
1.23.16.1	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.24.12.1	02-Aug-2025	perseant	Sync with HEAD
1.6	09-May-2016	christos	s/protected/libedit_private/g
1.5	12-Apr-2016	christos	From Ingo Schwarze: * Delete the stubs of the XK_EXE mechanism that was never implemented. From a security, stability, and simplicity perspective, i would consider implementing it a truly terrible idea, so let's better get rid of it. * Do not use the local variable "num" in el_wgets() alternately for two completely different purposes. Only use it for the number of characters read, as stated in the comment (or -1 as long as that number is still unknown), not for the (more or less boolean) return value of read_getcmd(). Actually, there is no need at all to save the latter return value after testing it once. * The function read_getcmd() has very unusual return values: It returns -1 for success and 0 for EOF/error. Switch that around to 0 for success and -1 for EOF/error to be less confusing, and get rid of the OKCMD preprocessor macro. * Get rid of one #ifdef section in el_wgets() by using el->el_chared.c_macro directly at the only place where it is used. * Delete the unused MIN() macro.
1.4	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.3	29-Jan-2016	christos	One macro is enough (Ingo Schwarze)
1.2	28-Jul-2011	christos	whitespace and comments
1.1	28-Jul-2011	christos	Rename key to keymacro to avoid conflicts with term.h. The renaming of term to terminal was again to avoid conflicts with term.h. term.h is a moving namespace violation.
1.2	21-Jun-2023	wiz	libedit: fix pkg-config to really provide readline directory as intended
1.1	20-Jun-2023	wiz	install pkg-config file for libedit version number matches portable libedit --cflags output matches portable libedit, since users probably want the readline interface
1.6	05-Dec-2024	christos	Don't eat 0 width characters, print them.
1.5	23-Jul-2019	christos	branches: 1.5.12; remove stray brace
1.4	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.3	30-Jun-2017	kre	branches: 1.3.4; 1.3.8; Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale).
1.2	29-Jun-2017	kre	Fix an obvious, but almost invisible typo (avoid some core dumps).
1.1	27-Jun-2017	christos	add literal escape sequence support, patterned after the tcsh ones.
1.3.8.1	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.3.4.2	23-Jul-2017	snj	Pull up following revision(s) (requested by kre in ticket #102): lib/libedit/Makefile: 1.64-1.65 lib/libedit/editline.3: 1.94-1.96 lib/libedit/editrc.5: 1.33 lib/libedit/el.c: 1.93-1.94 lib/libedit/el.h: 1.42 lib/libedit/literal.c: 1.1-1.3 lib/libedit/literal.h: 1.1-1.2 lib/libedit/prompt.c: 1.27 lib/libedit/read.c: 1.103 lib/libedit/refresh.c: 1.52-1.54 lib/libedit/refresh.h: 1.11 lib/libedit/terminal.c: 1.33 Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming. -- Include EDITRC in doc. -- mention the limitation of the literal sequence delimiter. -- - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display. -- - add literal sequence handling. -- remove unused variable -- add literal escape sequence support, patterned after the tcsh ones. -- fix comment -- Fix an obvious, but almost invisible typo (avoid some core dumps). -- Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale). -- Remove workaround for ancient HTML generation code.
1.3.4.1	30-Jun-2017	snj	file literal.c was added on branch netbsd-8 on 2017-07-23 14:41:26 +0000
1.5.12.1	02-Aug-2025	perseant	Sync with HEAD
1.2	30-Jun-2017	kre	branches: 1.2.4; Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale).
1.1	27-Jun-2017	christos	add literal escape sequence support, patterned after the tcsh ones.
1.2.4.2	23-Jul-2017	snj	Pull up following revision(s) (requested by kre in ticket #102): lib/libedit/Makefile: 1.64-1.65 lib/libedit/editline.3: 1.94-1.96 lib/libedit/editrc.5: 1.33 lib/libedit/el.c: 1.93-1.94 lib/libedit/el.h: 1.42 lib/libedit/literal.c: 1.1-1.3 lib/libedit/literal.h: 1.1-1.2 lib/libedit/prompt.c: 1.27 lib/libedit/read.c: 1.103 lib/libedit/refresh.c: 1.52-1.54 lib/libedit/refresh.h: 1.11 lib/libedit/terminal.c: 1.33 Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming. -- Include EDITRC in doc. -- mention the limitation of the literal sequence delimiter. -- - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display. -- - add literal sequence handling. -- remove unused variable -- add literal escape sequence support, patterned after the tcsh ones. -- fix comment -- Fix an obvious, but almost invisible typo (avoid some core dumps). -- Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale). -- Remove workaround for ancient HTML generation code.
1.2.4.1	30-Jun-2017	snj	file literal.h was added on branch netbsd-8 on 2017-07-23 14:41:26 +0000
1.29	09-May-2016	christos	s/protected/libedit_private/g
1.28	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.27	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.26	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.25	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.24	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.23	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.22	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.21	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.20	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.19	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.18	21-Mar-2012	matt	Use C89 functions definitions.
1.17	28-Jul-2011	christos	branches: 1.17.2; include config.h for all c files.
1.16	18-Apr-2010	christos	shame on solaris that is the last OS not supporting $()
1.15	15-Apr-2010	christos	From Jess Thrysoee - Fix wint_t to Int confusion
1.14	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.13	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.12	12-Feb-2009	sketch	More fixes for existing portability stuff.
1.11	22-Oct-2005	christos	branches: 1.11.28; change tr '[a-z]' '[A-Z]' to tr '[:lower:]' '[:upper:]' so that POSIX systems work properly regardless of environment variable settings.
1.10	08-Aug-2005	christos	Don't add an extra { NULL, 0, NULL } element to the help array. Instead keep it always the same size as the function array for consistency. Reported in FreeBSD PR 82381, but fixed differently.
1.9	16-May-2005	lukem	Remove clause 3 from the UCB license.
1.8	10-Mar-2003	christos	don't use the path for awk
1.7	09-Jan-2001	jdolecek	make array of functions and help array const
1.6	04-Sep-2000	lukem	- generate ansi prototypes instead of using __P(). noted by christos - fix a couple of comments
1.5	21-Jun-2000	lukem	* add -m option to makelist, which generates an mdoc table with the key bindings and their descriptions * manually add the output of 'sh ./makelist -m vi.c ed.c common.c' to a new section in editrc(5) called `EDITOR COMMANDS'
1.4	02-Jul-1999	simonb	branches: 1.4.8; More trailing white space.
1.3	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.2	07-Jun-1995	cgd	be a bit more careful when splitting pathnames
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.4.8.1	23-Jun-2000	lukem	Pull up editrc.5 revision 1.9 Pull up makelist revision 1.5 * add -m option to makelist, which generates an mdoc table with the key bindings and their descriptions * manually add the output of 'sh ./makelist -m vi.c ed.c common.c' to a new section in editrc(5) called `EDITOR COMMANDS' requested/approved by thorpej
1.11.28.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.17.2.1	17-Apr-2012	yamt	sync with head
1.56	03-Jan-2025	rillig	libedit: remove redundant break statements after EL_ABORT
1.55	30-Oct-2022	christos	branches: 1.55.4; improvements in malloc/free handling.
1.54	29-Aug-2021	christos	^W is traditionally bound to ed-delete-prev-word and not kill-region ^R is traditionally bound to em-inc-search-next and not redisplay This is what mksh, zsh bash, readline do (Baptiste Daroussin)
1.53	30-Mar-2020	ryo	fix build error with SDEBUG, MAP_DEBUG, DEBUG_REFRESH
1.52	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.51	09-May-2016	christos	branches: 1.51.16; s/protected/libedit_private/g
1.50	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.49	17-Apr-2016	christos	Remove empty callbacks (Ingo Schwartze)
1.48	12-Apr-2016	christos	From Ingo Schwarze: * Delete the stubs of the XK_EXE mechanism that was never implemented. From a security, stability, and simplicity perspective, i would consider implementing it a truly terrible idea, so let's better get rid of it. * Do not use the local variable "num" in el_wgets() alternately for two completely different purposes. Only use it for the number of characters read, as stated in the comment (or -1 as long as that number is still unknown), not for the (more or less boolean) return value of read_getcmd(). Actually, there is no need at all to save the latter return value after testing it once. * The function read_getcmd() has very unusual return values: It returns -1 for success and 0 for EOF/error. Switch that around to 0 for success and -1 for EOF/error to be less confusing, and get rid of the OKCMD preprocessor macro. * Get rid of one #ifdef section in el_wgets() by using el->el_chared.c_macro directly at the only place where it is used. * Delete the unused MIN() macro.
1.47	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.46	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.45	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.44	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.43	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.42	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.41	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.40	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.39	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.38	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.37	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.36	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.35	14-May-2015	christos	fix warnings on ubuntu 32 bit (Miki Rozloznik)
1.34	06-Jul-2014	christos	Bounds search for reallocated index, from OpenBSD via Andreas Fett
1.33	01-Jan-2013	christos	branches: 1.33.6; remove dead assignment (Christoph Mallon)
1.32	01-Jan-2013	christos	Fix pasto that affected bind -k (Christoph Mallon)
1.31	18-Nov-2011	christos	branches: 1.31.6; Add coverity annotations about unreachable code (Kamil Dudka)
1.30	16-Aug-2011	christos	branches: 1.30.2; re-enable -Wconversion
1.29	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.28	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.27	28-Jul-2011	christos	Rename key to keymacro to avoid conflicts with term.h. The renaming of term to terminal was again to avoid conflicts with term.h. term.h is a moving namespace violation.
1.26	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.25	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.24	09-Apr-2006	christos	fix debugging printf format.
1.23	06-Mar-2006	christos	Print the actual eofc, instead of ^D\b\b. Change internal character decoding to prevent buffer oveflows.
1.22	09-Aug-2005	christos	Fix two more help iterators. Thanks Stefan Farfeleder!
1.21	08-Aug-2005	christos	Don't add an extra { NULL, 0, NULL } element to the help array. Instead keep it always the same size as the function array for consistency. Reported in FreeBSD PR 82381, but fixed differently.
1.20	13-Aug-2004	mycroft	Delete-previous-char and delete-next-char without an argument are not supposed to modify the yank buffer in Emacs. Make it so.
1.19	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.18	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.17	31-Oct-2002	christos	support for % command [matching parens/brackets/braces] on vi modes. From David Laight, thanks!
1.16	27-Oct-2002	christos	vi mode and memory fixes from david laight.
1.15	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.14	09-Jan-2001	jdolecek	make constant arrays a const
1.13	04-Jan-2001	christos	consistently check for allocation failures and return -1, if we could not get more memory.
1.12	11-Nov-2000	christos	- add support for home and end keys. - improve debugging support
1.11	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.10	15-Oct-1999	jdolecek	don't assume locales are not working - it may not be the case re_refresh(): cast the character passed to re_addc() to unsigned char, so we don't end up calling isprint() with negative value when chars are signed and character value is >= 128
1.9	02-Jul-1999	simonb	branches: 1.9.2; More trailing white space.
1.8	12-Dec-1998	christos	delint
1.7	29-Jul-1998	lukem	* add more checks for NULL pointers in passed arguments * implement el_get(EditLine , int op, void result), which does the inverse of el_set() * add EL_EDITMODE operation to el_set and el_get; if non zero editing is enabled (the default). * add "edit on \| off" editrc command, which modifies EL_EDITMODE. users can now add '*:edit off' in ~/.editrc as an advisory to disable editing. NOTE: at this time EL_EDITMODE is just an indication of the state of the 'edit' command. It's up to the application to check this after el_source() or el_parse() to determine if editing is still required.
1.6	30-Mar-1998	mrg	use int rather than char as an array index.
1.5	03-Feb-1998	perry	remove obsolete register declarations
1.4	06-Jul-1997	christos	Fix compiler warnings.
1.3	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.2	27-Apr-1995	mycroft	Trivial code ordering change.
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.9.2.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.30.2.2	23-Jan-2013	yamt	sync with head
1.30.2.1	17-Apr-2012	yamt	sync with head
1.31.6.2	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.31.6.1	25-Feb-2013	tls	resync with head
1.33.6.1	10-Aug-2014	tls	Rebase.
1.51.16.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.51.16.1	08-Apr-2020	martin	Merge changes from current as of 20200406
1.55.4.1	02-Aug-2025	perseant	Sync with HEAD
1.13	09-May-2016	christos	s/protected/libedit_private/g
1.12	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.11	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.10	06-Jul-2014	christos	Bounds search for reallocated index, from OpenBSD via Andreas Fett
1.9	30-Dec-2009	christos	branches: 1.9.12; 1.9.22; Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.8	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.7	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.6	09-Jan-2001	jdolecek	make constant arrays a const
1.5	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.4	02-Jul-1999	simonb	More trailing white space.
1.3	29-Jul-1998	lukem	* add more checks for NULL pointers in passed arguments * implement el_get(EditLine , int op, void result), which does the inverse of el_set() * add EL_EDITMODE operation to el_set and el_get; if non zero editing is enabled (the default). * add "edit on \| off" editrc command, which modifies EL_EDITMODE. users can now add '*:edit off' in ~/.editrc as an advisory to disable editing. NOTE: at this time EL_EDITMODE is just an indication of the state of the 'edit' command. It's up to the application to check this after el_source() or el_parse() to determine if editing is still required.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.9.22.1	10-Aug-2014	tls	Rebase.
1.9.12.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.42	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.41	29-Nov-2018	christos	Fix off by one <tsahara at iij>
1.40	09-May-2016	christos	branches: 1.40.14; 1.40.16; s/protected/libedit_private/g
1.39	11-Apr-2016	christos	Fix indentation, Ingo Schwarze
1.38	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.37	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.36	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.35	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.34	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.33	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.32	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.31	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.30	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.29	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.28	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.27	06-Jul-2014	christos	Bounds search for reallocated index, from OpenBSD via Andreas Fett
1.26	16-Aug-2011	christos	branches: 1.26.8; 1.26.18; re-enable -Wconversion
1.25	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.24	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.23	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.22	29-May-2005	lukem	Update for recent parse__escape() prototype change
1.21	29-May-2005	christos	PR/25694: Luke Mewburn: Don't abuse unconstify'ing a string and writing to it, because you'll core dump. Also remove extra const that gives pain to the irix compiler.
1.20	05-Dec-2003	lukem	Tokenization function enhancements: * Make tok_init(), tok_end(), tok_reset(), tok_line() and tok_str() publically available in <histedit.h> * Documented the public functions in editline(3) * Renamed tok_line() -> tok_str() * Added new tok_line() which takes a "const LineInfo " instead of "const char " (the former has "cursor" information), and optionally return the argv index ("int cursorc") and offset within that index ("int cursorv"). This means that completion routines can use the tokenization code to crack the line and easily find which word the cursor is at. (mmm, context sensitive completion :) * Fixed TEST/test.c when using "continuation" lines (unmatched quote or \ at EOL), and added some more DEBUG messages including highlighting where the cursor is (with a `_').
1.19	02-Nov-2003	christos	Handle M- as escape. XXX: should probably select the meta-map instead. From Gerry Swislow gerry at certif com
1.18	15-Oct-2003	christos	don't limit ^c to alpha c, and add VIS_NOSLASH so that vis(3) does not produce \^c
1.17	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.16	21-Jan-2003	christos	Add a uniquefier for the history function.
1.15	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.14	23-Jan-2001	jdolecek	sprinkle couple const
1.13	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.12	10-Mar-2000	jdolecek	include <stdlib.h> to get definition of malloc() and free(), so it's possible to compile this file separately when debugging
1.11	02-Jul-1999	simonb	More trailing white space.
1.10	05-Feb-1999	christos	M-X:<enter> core-dumped.
1.9	12-Dec-1998	christos	delint
1.8	29-Jul-1998	lukem	* add more checks for NULL pointers in passed arguments * implement el_get(EditLine , int op, void result), which does the inverse of el_set() * add EL_EDITMODE operation to el_set and el_get; if non zero editing is enabled (the default). * add "edit on \| off" editrc command, which modifies EL_EDITMODE. users can now add '*:edit off' in ~/.editrc as an advisory to disable editing. NOTE: at this time EL_EDITMODE is just an indication of the state of the 'edit' command. It's up to the application to check this after el_source() or el_parse() to determine if editing is still required.
1.7	21-Jan-1998	lukem	in el_parse(), use a temporary buffer to store the program name when comparing, preventing trashing of argv[0]. remove note in man page warning of former behaviour.
1.6	06-Jul-1997	christos	Fix compiler warnings.
1.5	11-Jan-1997	lukem	* document ^char and \ escape sequences * when parsing ^char control chars, check the correct char when determining validity (previously, ^char was a NOP interpreted as the literal string because of this bug)
1.4	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.3	11-Jan-1997	lukem	* in el_parse(), don't reference argv[0] if argc < 1 (return -1 instead) * clarify return value of el_parse()
1.2	09-Jan-1997	lukem	* add a man page for the editline routines * add a man page describing editrc * fix bugs in el_parse(): * didn't execute command when program name matched (test reversed) * was checking against empty string instead of program name * after checks, command to run also pointed to empty string [christos - the author of libedit - ok-ed the man pages in general (which I wrote from scratch by RTFS) as well as the bugfix]
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.26.18.1	10-Aug-2014	tls	Rebase.
1.26.8.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.40.16.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.40.16.1	10-Jun-2019	christos	Sync with HEAD
1.40.14.1	26-Dec-2018	pgoyette	Sync with HEAD, resolve a few conflicts
1.9	09-May-2016	christos	s/protected/libedit_private/g
1.8	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.7	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.6	29-May-2005	lukem	Update for recent parse__escape() prototype change
1.5	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.4	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.3	02-Jul-1999	simonb	More trailing white space.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.27	27-Jun-2017	christos	add literal escape sequence support, patterned after the tcsh ones.
1.26	09-May-2016	christos	branches: 1.26.8; s/protected/libedit_private/g
1.25	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.24	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.23	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.22	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.21	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.20	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.19	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.18	31-Dec-2009	christos	- Document and enable wide character support. - Fix read function compatibility.
1.17	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.16	17-Jul-2009	christos	handle prompt_esc properly.
1.15	16-Apr-2009	christos	PR/41230: -current: sh(1) endlessly looping in interactive Fix proposed from Matthew Mondor
1.14	31-Mar-2009	christos	Implement literal prompt sequences. Now someone can implement RL_PROMPT_START_LITERAL/RL_PROMPT_END_LITERAL :-)
1.13	21-Feb-2009	christos	back out all prompt changes. they are not needed.
1.12	17-Feb-2009	christos	allow for a prompt argument.
1.11	07-Aug-2003	agc	branches: 1.11.40; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.10	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.9	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.8	10-Jan-2001	jdolecek	Enlarge editline buffers as needed to support arbitrary length lines. This also addresses lib/9712 by Phil Nelson.
1.7	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.6	12-Nov-1999	lukem	- implement printing a right-side prompt. code derived from similar work I wrote for tcsh(1) three years ago. - implement EL_RPROMPT, which allows a setting/getting of a function which returns a string to be used as the right-side prompt. - improve HISTORY and AUTHORS sections in editline(3). - bump shlib minor version for EL_RPROMPT. XXX: due to an implementation issue, the rprompt has a 1 space gap before the edge of the logical screen. editline's logical screen is 1 space less than the full screen width, so there's a 2 space gap between the rprompt and the right end of the physical screen. i'm not concerned about this.
1.5	02-Jul-1999	simonb	branches: 1.5.4; More trailing white space.
1.4	29-Jul-1998	lukem	* add more checks for NULL pointers in passed arguments * implement el_get(EditLine , int op, void result), which does the inverse of el_set() * add EL_EDITMODE operation to el_set and el_get; if non zero editing is enabled (the default). * add "edit on \| off" editrc command, which modifies EL_EDITMODE. users can now add '*:edit off' in ~/.editrc as an advisory to disable editing. NOTE: at this time EL_EDITMODE is just an indication of the state of the 'edit' command. It's up to the application to check this after el_source() or el_parse() to determine if editing is still required.
1.3	06-Jul-1997	christos	Fix compiler warnings.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.5.4.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.11.40.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.26.8.1	23-Jul-2017	snj	Pull up following revision(s) (requested by kre in ticket #102): lib/libedit/Makefile: 1.64-1.65 lib/libedit/editline.3: 1.94-1.96 lib/libedit/editrc.5: 1.33 lib/libedit/el.c: 1.93-1.94 lib/libedit/el.h: 1.42 lib/libedit/literal.c: 1.1-1.3 lib/libedit/literal.h: 1.1-1.2 lib/libedit/prompt.c: 1.27 lib/libedit/read.c: 1.103 lib/libedit/refresh.c: 1.52-1.54 lib/libedit/refresh.h: 1.11 lib/libedit/terminal.c: 1.33 Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming. -- Include EDITRC in doc. -- mention the limitation of the literal sequence delimiter. -- - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display. -- - add literal sequence handling. -- remove unused variable -- add literal escape sequence support, patterned after the tcsh ones. -- fix comment -- Fix an obvious, but almost invisible typo (avoid some core dumps). -- Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale). -- Remove workaround for ancient HTML generation code.
1.15	09-May-2016	christos	s/protected/libedit_private/g
1.14	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.13	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.12	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.11	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.10	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.9	31-Mar-2009	christos	Implement literal prompt sequences. Now someone can implement RL_PROMPT_START_LITERAL/RL_PROMPT_END_LITERAL :-)
1.8	21-Feb-2009	christos	back out all prompt changes. they are not needed.
1.7	17-Feb-2009	christos	allow for a prompt argument.
1.6	07-Aug-2003	agc	branches: 1.6.40; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.5	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.4	12-Nov-1999	lukem	- implement printing a right-side prompt. code derived from similar work I wrote for tcsh(1) three years ago. - implement EL_RPROMPT, which allows a setting/getting of a function which returns a string to be used as the right-side prompt. - improve HISTORY and AUTHORS sections in editline(3). - bump shlib minor version for EL_RPROMPT. XXX: due to an implementation issue, the rprompt has a 1 space gap before the edge of the logical screen. editline's logical screen is 1 space less than the full screen width, so there's a 2 space gap between the rprompt and the right end of the physical screen. i'm not concerned about this.
1.3	29-Jul-1998	lukem	branches: 1.3.6; * add more checks for NULL pointers in passed arguments * implement el_get(EditLine , int op, void result), which does the inverse of el_set() * add EL_EDITMODE operation to el_set and el_get; if non zero editing is enabled (the default). * add "edit on \| off" editrc command, which modifies EL_EDITMODE. users can now add '*:edit off' in ~/.editrc as an advisory to disable editing. NOTE: at this time EL_EDITMODE is just an indication of the state of the 'edit' command. It's up to the application to check this after el_source() or el_parse() to determine if editing is still required.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.3.6.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.6.40.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.109	03-Jan-2025	rillig	libedit: remove redundant break statements after EL_ABORT
1.108	30-Oct-2022	christos	branches: 1.108.4; improvements in malloc/free handling.
1.107	15-Aug-2021	christos	Disable attempts to handle EINTR and non-blocking I/O by default. It is confusing to other programs and unexpected behavior. Reported by Ingo Schwarze. This behavior is now controlled with EL_SAFEREAD.
1.106	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.105	25-Nov-2018	christos	From Yuichiro Naito (FreeBSD): hrs@ says that (cbp >= MB_LEN_MAX) condition is necessary for checking invalid byte sequences. If malicious input was given, libedit would read byte sequences forever.
1.104	18-Nov-2018	christos	Remove utf-8 requirement (Yuichiro NAITO)
1.103	27-Jun-2017	christos	branches: 1.103.4; 1.103.6; remove unused variable
1.102	11-Dec-2016	christos	branches: 1.102.6; PR/51706: Amir Plivatsky: Fix memory leak
1.101	25-May-2016	christos	branches: 1.101.2; abstract read code to a single function (Ingo Schwarze)
1.100	24-May-2016	christos	el_map.alt can't be NULL here (Ingo Schwarze)
1.99	24-May-2016	christos	remove debug read (Ingo Schwarze)
1.98	24-May-2016	christos	From Ingo Schwarze: Reduce obfuscation of errno handling. There is only one purpose non-local errno handling is needed for: Inside el_wgets(), several functions call down indirectly to el_wgetc(), many of them via the dispatch table. When el_wgetc() fails, it does properly report failure, but then various cleanup is done which may clobber errno. But when returning due to failure, el_wgets() wants to have errno set to the reason of the original read failure, not to the reason of some subsequent failure of some cleanup operation. So el_wgetc() needs to save errno, and if it's non-zero, el_wgets() needs to restore it on failure. This core logic is currently obscured by the fact that el_errno is set and inspected at some additional places where it isn't needed. Besides, since el_wgetc() and and el_wgets() are both in read.c, el_errno does not need to be in struct editline, it can and should be local to read.c in struct el_read_t. Let's look at what can be simplified. 1. keymacro_get() abuses el_errno instead of having a proper error return code. Adding that error return code is easy because node_trav() already detects the condition and an adequate code is already defined. Returning it, testing for it in read_getcmd(), and returning with error from there removes the need to inspect el_errno from el_wgets() after calling read_getcmd(). Note that resetting lastchar and cursor and clearing buffer[0] is irrelevant. The code returns from el_wgets() right afterwards. Outside el_wgets(), these variables are no longer relevant. When el_wgets() is called the next time, it will call ch_reset() anyway, resetting the two pointers. And as long as lastchar points to the beginning of the buffer, the contents of the buffer won't be used for anything. 2. read_getcmd() doesn't need to set el_errno again after el_wgetc() failure since el_wgetc() already did so. While here, remove the silly "if EOF or error" comments from the el_wgetc() return value tests. It's a public interface documented in a manual, so people working on the implementation can obviously be expected to know how it works. It's a case of count++; /* Increment count. / 3. In the two code paths of el_wgets() that lead up to "goto noedit", there is no need to save the errno because nothing that might change it happens before returning. For clarity, since el_wgets() is the function restoring the errno, also move initializing it to the same function. Finally, note that restoring errno when the saved value is zero is wrong. No library code is ever allowed to clear a previously set value of errno. Only application programs are allowed to do that, and even they usually don't need to do so, except when using certain ill-designed interfaces like strtol(3). I tested that the behaviour remains sane in the following cases, all during execution of el_wgets(3) and with a signal handler for USR1 installed without SA_RESTART. Enter some text and maybe move around a bit. Then send a USR1 signal. The signal gets processed, then read_char() resumes reading. Send another USR1 signal. Now el_wgets() sets errno=EINTR and returns -1. * Press Ctrl-V to activate ed-quoted-insert. Then send a USR1 signal. The signal gets processed, then read_char() resumes reading. Send another USR1 signal. ed_quoted_insert() returns ed_end_of_file(), i.e. CC_EOF, and el_wgets() returns 0. * Press a key starting a keyboard macro. Then send a USR1 signal. The signal gets processed, then read_char() resumes reading. Send another USR1 signal. Now el_wgets() sets errno=EINTR and returns -1. * Press : to enter builtin command mode. Start typing a command. Then send a USR1 signal. The signal gets processed, then read_char() resumes reading. Send another USR1 signal. Now c_gets() returns -1, ed_command() beeps and returns CC_REFRESH, and el_wgets() resumes operation as it should. I also tested with "el_set(el, EL_EDITMODE, 0)", and it returns the right value and sets errno correctly.
1.97	22-May-2016	christos	Stop the read module from poking the el_chared.c_macro data structure currently belonging to the chared module. The read module does so from three of its functions, while no other module uses the macro data, not even the chared module itself. That's quite logical because macros are a feature of input handling, all of which is done by the read module, and none by the chared module. So move the data into the read modules's own opaque data structure, struct el_read_t. That simplifies internal interfaces in several respects: The semi-public chared.h has one fewer struct, one fewer #define, and one fewer member in struct el_chared_t; all three move to one single C file, read.c, and are now module-local. And the internal interface function ch_reset() needs one fewer argument, making the code of many functions in various modules more readable. The price is one additional internal interface function, read_end(), 10 lines long including comments, called publicly from exactly one place: el_end() in el.c. That's hardly an increase in complexity since most other modules already have their *_end() function, read.c was the odd one out not having one. From Ingo Schwarze
1.96	09-May-2016	christos	s/protected/libedit_private/g
1.95	19-Apr-2016	christos	From Ingo Schwarze: - Put the data type el_rfunc_t into the public header <histedit.h>. - Make el_read in struct editline an opaque pointer rather than an embedded struct. - Do not include "read.h" everywhere, but only in the two files needing access to el_read, read.c and el.c. - To functions that don't need more, pass the struct el_read_t * rather than the full EditLine *. - Of course, that means that read_init() can now fail from memory exhaustion, but it's easy to clean up after that.
1.94	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.93	12-Apr-2016	christos	FIONREAD takes int as an argument (Ingo Schwarze)
1.92	12-Apr-2016	christos	From Ingo Schwarze: * Delete the stubs of the XK_EXE mechanism that was never implemented. From a security, stability, and simplicity perspective, i would consider implementing it a truly terrible idea, so let's better get rid of it. * Do not use the local variable "num" in el_wgets() alternately for two completely different purposes. Only use it for the number of characters read, as stated in the comment (or -1 as long as that number is still unknown), not for the (more or less boolean) return value of read_getcmd(). Actually, there is no need at all to save the latter return value after testing it once. * The function read_getcmd() has very unusual return values: It returns -1 for success and 0 for EOF/error. Switch that around to 0 for success and -1 for EOF/error to be less confusing, and get rid of the OKCMD preprocessor macro. * Get rid of one #ifdef section in el_wgets() by using el->el_chared.c_macro directly at the only place where it is used. * Delete the unused MIN() macro.
1.91	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.90	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.89	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.88	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.87	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.86	02-Mar-2016	christos	PR/50880: David Binderman: Remove redundant code. While here, fix all debugging formats.
1.85	24-Feb-2016	christos	Tuck in mbstate_t to the wide char version only to avoid exposing the zeroing hack and doing it in the narrow case.
1.84	24-Feb-2016	christos	Make the read_char function always take a wchar_t * argument (Ingo Schwarze)
1.83	24-Feb-2016	christos	Get split el_getc and el_wgetc completely and call el_wgetc internally. Change some character constants to they wide versions. (Ingo Schwarze)
1.82	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.81	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.80	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.79	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.78	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.77	14-Feb-2016	christos	From Ingo Schwartze: Next step: Remove #ifdef'ing in read_char(), in the same style as we did for setlocale(3) in el.c. A few remarks are required to explain the choices made. * On first sight, handling mbrtowc(3) seems a bit less trivial than handling setlocale(3) because its prototype uses the data type mbstate_t from <wchar.h>. However, it turns out that "histedit.h" already includes <wchar.h> unconditionally (i don't like headers including other headers, but that ship has sailed, people are by now certainly used to the fact that including "histedit.h" doesn't require including <wchar.h> before), and "histedit.h" is of course included all over the place. So from that perspective, there is no problem with using mbrtowc(3) unconditionally ever for !WIDECHAR. * However, <wchar.h> also defines the mbrtowc(3) prototype, so we cannot just #define mbrtowc away, or including the header will break. It would also be a bad idea to porovide a local implementation of mbrtowc() and hope that it overrides the one in libc. Besides, the required prototype is subtly different: While mbrtowc(3) takes "wchar_t " as its first argument, we need a function that takes "Char ". So unfortunately, we have to keep a ct_mbrtowc #define, at least until we can maybe get rid of "Char " in the more remote future. After getting rid of the #else clause in read_char(), we can pull "return 1;" into the default: clause. After that, we can get rid of the ugly "goto again_lastbyte;" and just "break;". As a bonus, that also gets rid of the ugly CONSTCOND. * While here, delete the unused ct_mbtowc() from chartype.h.
1.76	12-Feb-2016	christos	GC IGNORE_EXTCHARS and simplify code (Ingo Schwarze)
1.75	12-Feb-2016	christos	From Ingo Schwarze: If CHARSET_IS_UTF8 is not set, read_char() is broken in a large number of ways: 1. The isascii(3) check can yield false positives. If a string in an arbitrary encoding contains a byte in the range 0..127, that does not at all imply that it forms a character all by itself, and even less that it represents the same character as in ASCII. Consequently, read_char() may return characters the user never typed. Even if the encoding is not state dependent, the assumption that bytes in the range 0..127 represent ASCII characters is broken. Consider UTF-16, for example. 2. The reverse problem can also occur. In an arbitrary encoding, there is no guarantee that a character that can be represented by ASCII is represented by a seven-bit byte, and even less by the same byte as in ASCII. Even for single-byte encodings, these assumptions are broken. Consider the ISO 646 national variants, for example. Consequently, the current code is insufficient to keep ASCII characters working even for single-byte encodings. 3. The condition "++cbp != 1" can never trigger (because initially, cbp is 0, and the code can only go back up via the final goto, which has another cbp = 0 right before it) and it has no effect (because cbp isn't used afterwards). 4. bytes = ct_mbtowc(cp, cbuf, cbp) is broken. If this returns -1, the code assumes that is can just call mbtowc(3) again for later input bytes. In some implementations, that may even be broken for state-independent encodings, but trying again after mbtowc(3) failure certainly produces completely erratic and meaningless results in state-dependent encodings. 5. The assignment "cp = (Char)(unsigned char)cbuf[0]" is completely bogus. Even if the byte cbuf[0] represents a character all by itself, which it usually will not, whether or not the cast produces the desired result depends on the internal representation of wchar_t in the C library, which the application program can know nothing about. Even for ASCII in the C/POSIX locale, an ASCII character other than '\0' == L'\0' == 0 need not have the same numeric value as a char and as a wchar_t. To summarize, this code only works if all of the following conditions hold: - The encoding is a single-byte encoding. - ASCII is a subset of the encoding. - The implementation of mbtowc(3) in the C library does not require re-initialization after encoding errors. - The implementation of wchar_t in the C library uses the same numerical values as ASCII. Otherwise, it silently produces wrong results. The simplest way to fix this is to just use the same code as for UTF-8 (right above). Of course, that causes functional changes but that shouldn't matter since current behaviour is undefined. The patch below provides the following improvements: - It works for all stateless single-byte encodings, no matter whether they are somehow related to ASCII, no matter how mb[r]towc(3) are internally implemented, and no matter how wchar_t is internally represented. - Instead of producing unpredictable and definitely wrong results for non-UTF-8 multibyte characters, it behaves in a well-defined way: It aborts input processing, sets errno, and returns failure. Note that short of providing full support for arbitrary locales, it is impossible to do better. We cannot know whether a given unsupported locale is state-dependent, and for a state-dependent locale, it makes no sense to retry parsing after an encoding error, so the best we can do is abort processing for any* unsupported multi-byte character. - Note that single-byte characters in arbitrary state-independent locales still work, even in locales that may potentially also contain multibyte characters, as long as those don't occur in input. I'm not sure whether any such locales exist in practice... Tested with UTF-8 and C/POSIX on OpenBSD. Also tested that in the C/POSIX locale, non-ASCII bytes get through unmangled. You may wish to test with ISO-LATIN on NetBSD if NetBSD supports that. ---- Also use a constant for meta to avoid warnings.
1.74	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.73	11-Feb-2016	christos	Remove utf8_islead() mbrtowc() handles this just fine (Ingo Schwarze)
1.72	08-Feb-2016	christos	UTF-8 fixes from Ingo Schwarze: 1. Assume that errno is non-zero when entering read_char() and that read(2) returns 0 (indicating end of file). Then, the code will clear errno before returning. (Obviously, the statement "errno = 0" is almost always a bug unless there is save_errno = errno right before it and the previous value is properly restored later, in all reachable code paths.) 2. When encountering an invalid byte sequence, the code discards all following bytes until MB_LEN_MAX overflows; consider, for example, 0xc2 immediately followed by a few valid ASCII bytes. Three of those ASCII bytes will be discarded. 3. On a POSIX system, EILSEQ will always be set after reading a valid (yes, valid, not invalid!) UTF-8 character. The reason is that mbtowc(3) will first be called with a length limit (third argument) of 1, which will fail, return -1, and - on a POSIX system - set errno to EILSEQ. This third bug is mitigated a bit because i couldn't find any system that actually conforms to POSIX in this respect: None of OpenBSD, NetBSD, FreeBSD, Solaris 11, and glibc set errno when an incomplete character is passed to mbtowc(3), even though that is required by POSIX. Anyway, that mbtowc(3) bug will be fixed at least in OpenBSD after release unlock, so it would be good to fix this bug in libedit before fixing the bug in mbtowc(3). How can these three bugs be fixed? 1. As far as i understand it, the intention of the bogus errno = 0 is to undo the effects of failing system calls in el_wset(), sig_set(), and read__fixio() if the subsequent read(2) indicates end of file. So, restoring errno has to be moved right after read__fixio(). Of course, neither 0 nor e is the right value to restore: 0 is wrong if errno happened to be set on entry, e would be wrong because if one read(2) fails but a second attempt succeeds after read__fixio(), errno should not be touched. So, the errno to be restored in this case has to be saved before calling read(2) for the first time. 2. Solving the second issue requires distinguishing invalid and incomplete characters, but that is impossible with the function mbtowc(3) because it returns -1 in both cases and sets errno to EILSEQ in both cases (once properly implemented). It is vital that each input character is processed right away. It is not acceptable to wait for the next input character before processing the previous one because this is an interactive library, not a batch system. Consequently, the only situation where it is acceptable to wait for the next byte without first processing the previous one(s) is when the previous one(s) form an incomplete sequence that can be continued to form a valid character. Consequently, short of reimplementing a full UTF-8 state machine by hand, the only correct way forward is to use mbrtowc(3). Even then, care is needed to always have the state object properly initialized before using it, and to not discard a valid ASCII or UTF-8 lead byte if it happens to follow an invalid sequence. 3. Fortunately, solution 2. also solves issue 3. as a side effect, by no longer using mbtowc(3) in the first place.
1.71	06-Jul-2014	christos	Bounds search for reallocated index, from OpenBSD via Andreas Fett
1.70	27-May-2013	christos	branches: 1.70.4; Test early for EOF to avoid infinite loop in the wide char case. From Linas Vepstas: linasvepstas at gmail dot com
1.69	11-Sep-2012	christos	branches: 1.69.2; return !OKCMD on error.
1.68	10-Sep-2012	christos	PR/46935: Steffen Nurpmeso: editline(3) (libedit): faulty errno handling, faulty reuse of val in wrong context
1.67	16-Aug-2011	christos	branches: 1.67.2; re-enable -Wconversion
1.66	29-Jul-2011	christos	pass -Wconversion
1.65	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.64	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.63	28-Jul-2011	christos	Rename key to keymacro to avoid conflicts with term.h. The renaming of term to terminal was again to avoid conflicts with term.h. term.h is a moving namespace violation.
1.62	28-Jul-2011	christos	term -> terminal fix wide function confusion
1.61	09-Jul-2011	christos	make obvious what we are trying to do...
1.60	09-Jul-2011	tron	Try to fix build of port where "char" is unsigned.
1.59	08-Jul-2011	christos	Support other non-latin1 single byte character sets. From: Alexander Barkov and Nirbhay Choubey at oracle dot com
1.58	18-Feb-2011	christos	PR/44599: Steven Vernon: libedit acts as if no data read if editmode is turned off because nread is uninitialized in the edit disabled case.
1.57	21-Jul-2010	christos	branches: 1.57.2; refresh only on SIGCONT not SIGWINCH from Edward Sheldrake
1.56	19-Jul-2010	christos	retry the read after sigwinch too, from Edward Sheldrake
1.55	22-Mar-2010	christos	https://bugzilla.redhat.com/show_bug.cgi?id=575383 Handle EINTR properly.
1.54	31-Dec-2009	christos	- Document and enable wide character support. - Fix read function compatibility.
1.53	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.52	22-Jul-2009	christos	Always initialize nread since it is an out param. From Michael Cook mcook at bbn dot com
1.51	09-Jun-2009	christos	decrement the number of levels after the loop (Julien Torres)
1.50	08-Jun-2009	christos	from Julien Torres, flip the order we pop in the macro array.
1.49	10-Mar-2009	christos	make el_gets set the count to -1 on error to distinguish between EOF and error.
1.48	21-Feb-2009	christos	remove VEOF test. the tty is in cooked mode when we are not editing and the tty driver does the check for us.
1.47	21-Feb-2009	christos	use the VEOF character from the terminal, instead of hard-coding 4.
1.46	19-Feb-2009	christos	reset and redraw on sigcont. From Anon Ymous.
1.45	15-Feb-2009	christos	pass lint on _LP64.
1.44	15-Feb-2009	christos	don't restart on EINTR, instead return NULL immediately. From Anon Ymous
1.43	05-Feb-2009	christos	branches: 1.43.2; match documentation in el_push
1.42	18-Jan-2009	lukem	fix -Wsign-compare issues
1.41	10-Sep-2008	christos	Allow a single process to control multiple ttys (for pthreads using _REENTRANT) using multiple EditLine objects. Mostly from Preston A. Elder.
1.40	01-Mar-2007	christos	branches: 1.40.12; Fix bug with multiple pending el_pushes. Reported by Julien Torres.
1.39	02-Aug-2005	christos	On a fatal error, we want to stop processing the macro buffers.
1.38	02-Aug-2005	tron	Add missing second argument to another call of ch_reset().
1.37	01-Aug-2005	christos	Don't reset the macro strings each time we enter el_gets(), otherwise el_push() is unusable programmatically.
1.36	01-Jun-2005	lukem	Don't use non-standard uint or u_int.
1.35	09-Mar-2005	christos	Make sure we flush after we prepare when we are unbuffered otherwise the prompt will not appear immediately.
1.34	08-Jul-2004	christos	PR/23486: Andreas Gustafsson: gdb no longer works with emacs - make sure that we keep previous contents of the buffer in unbuffered mode. - when turning editing on and off keep tty consistent.
1.33	27-Feb-2004	christos	branches: 1.33.2; Better fix for rl_prep_terminal() from Gerry Swislow.
1.32	21-Feb-2004	christos	Separate tty separation from the prompt printing function. From Gerry Swislow.
1.31	17-Jan-2004	christos	portability fixes.
1.30	18-Oct-2003	christos	change allocation policy in el_push to allocate the string itself. fix issues with strdup.
1.29	09-Oct-2003	christos	PR/23107: Nathan Williams: ^D as the first char on the command line does not DTRT in readline compatibility mode
1.28	26-Sep-2003	christos	Implement enough of readline's 4.0 async mode to make gdb happy. This is not complete yet, but it seems to work... This required to introduce an unbuffered mode to el_gets(), but that was a minor change.
1.27	13-Sep-2003	mycroft	Fix something that's been annoying me for a while... Pull in <fcntl.h>, so that the fcntl #defines are present, and we build in the automatic reset of non-blocking mode, rather than beeping like mad.
1.26	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.25	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.24	20-Nov-2002	christos	Fix problem with previous patches that broke vi history. - c_gets() was usually returning a length, but sometimes one of the CC_xxx values (which are small +ve integers)! - fixed c_gets() by putting a ' ' under the cursor. From David Laight.
1.23	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.22	27-Oct-2002	christos	vi mode and memory fixes from david laight.
1.21	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.20	27-Sep-2001	christos	PR/14067: Anthony Mallet: Provide a programmatic way to set the read_char function via a new el_set() operation. Thanks, nicely done :-)
1.19	10-Jan-2001	jdolecek	Enlarge editline buffers as needed to support arbitrary length lines. This also addresses lib/9712 by Phil Nelson.
1.18	11-Nov-2000	christos	- add support for home and end keys. - improve debugging support
1.17	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.16	19-Jan-2000	christos	PR/9243: Kevin Schoedel: libedit ignores repeat count
1.15	04-Oct-1999	lukem	remove some lint
1.14	08-Aug-1999	sommerfeld	branches: 1.14.2; minor tweak to previous fix: don't spuriously truncate final newline under emacs.
1.13	02-Aug-1999	sommerfeld	Fix PR7685 (gdb under emacs prints spurious ^M and messes up terminal) plus a few bogons noted along the way: 1) Set EDIT_DISABLED if terminal type is emacs. 2) fix bug in NO_TTY mode which caused it to not notice CR or LF 3) implement EDIT_DISABLED within libedit to be somewhat like NO_TTY, except that a prompt is printed first.
1.12	02-Jul-1999	simonb	More trailing white space.
1.11	11-Jan-1999	kleink	In userland, pull in <errno.h> instead of <sys/errno.h> for the declaration of errno.
1.10	12-Dec-1998	christos	delint
1.9	01-Jun-1998	lukem	* implement CC_REFRESH_BEEP; as per CC_REFRESH but beep as well. this is useful in completion when a partial completion is found * remove entry in BUGS about el_parse(); that was fixed a while ago
1.8	21-Jan-1998	lukem	fix compile errors if FIONREAD is defined. noted by David Holland <dholland@bordeaux.eecs.harvard.edu> in [bin/4012].
1.7	20-Dec-1997	christos	Small optimization. Don't call isatty() on every invocation of el_gets, but remember if the tty setup failed... Also trim the input line of trailing \r's.
1.6	26-Oct-1997	christos	Make el_gets() work when the input is not a tty.
1.5	06-Jul-1997	christos	Fix compiler warnings.
1.4	11-Apr-1997	christos	Portability fixes: __const -> const BADSIG -> SIG_ERR int flags -> u_int flags #if __STDC__ -> #ifdef __STDC__
1.3	14-Jan-1997	lukem	Implement CC_REDISPLAY, which (unlink CC_REFRESH) redraws the entire input line (a la ^R). This is useful if the binding outputs information and mucks up the input line. To be used in ``list-choices'' bindings (refer to the ^D binding in csh when filec is set)
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.14.2.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.33.2.1	10-Jul-2004	tron	Pull up revision 1.34 (requested by christos in ticket #620): PR/23486: Andreas Gustafsson: gdb no longer works with emacs - make sure that we keep previous contents of the buffer in unbuffered mode. - when turning editing on and off keep tty consistent.
1.40.12.1	24-Sep-2008	wrstuden	Merge in changes between wrstuden-revivesa-base-2 and wrstuden-revivesa-base-3.
1.43.2.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.57.2.1	05-Mar-2011	bouyer	Sync with HEAD
1.67.2.2	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.67.2.1	30-Oct-2012	yamt	sync with head
1.69.2.2	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.69.2.1	23-Jun-2013	tls	resync from head
1.70.4.1	10-Aug-2014	tls	Rebase.
1.101.2.1	07-Jan-2017	pgoyette	Sync with HEAD. (Note that most of these changes are simply $NetBSD$ tag issues.)
1.102.6.1	23-Jul-2017	snj	Pull up following revision(s) (requested by kre in ticket #102): lib/libedit/Makefile: 1.64-1.65 lib/libedit/editline.3: 1.94-1.96 lib/libedit/editrc.5: 1.33 lib/libedit/el.c: 1.93-1.94 lib/libedit/el.h: 1.42 lib/libedit/literal.c: 1.1-1.3 lib/libedit/literal.h: 1.1-1.2 lib/libedit/prompt.c: 1.27 lib/libedit/read.c: 1.103 lib/libedit/refresh.c: 1.52-1.54 lib/libedit/refresh.h: 1.11 lib/libedit/terminal.c: 1.33 Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming. -- Include EDITRC in doc. -- mention the limitation of the literal sequence delimiter. -- - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display. -- - add literal sequence handling. -- remove unused variable -- add literal escape sequence support, patterned after the tcsh ones. -- fix comment -- Fix an obvious, but almost invisible typo (avoid some core dumps). -- Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale). -- Remove workaround for ancient HTML generation code.
1.103.6.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.103.6.1	10-Jun-2019	christos	Sync with HEAD
1.103.4.1	26-Nov-2018	pgoyette	Sync with HEAD, resolve a couple of conflicts
1.108.4.1	02-Aug-2025	perseant	Sync with HEAD
1.13	30-Oct-2022	christos	improvements in malloc/free handling.
1.12	22-May-2016	christos	Stop the read module from poking the el_chared.c_macro data structure currently belonging to the chared module. The read module does so from three of its functions, while no other module uses the macro data, not even the chared module itself. That's quite logical because macros are a feature of input handling, all of which is done by the read module, and none by the chared module. So move the data into the read modules's own opaque data structure, struct el_read_t. That simplifies internal interfaces in several respects: The semi-public chared.h has one fewer struct, one fewer #define, and one fewer member in struct el_chared_t; all three move to one single C file, read.c, and are now module-local. And the internal interface function ch_reset() needs one fewer argument, making the code of many functions in various modules more readable. The price is one additional internal interface function, read_end(), 10 lines long including comments, called publicly from exactly one place: el_end() in el.c. That's hardly an increase in complexity since most other modules already have their *_end() function, read.c was the odd one out not having one. From Ingo Schwarze
1.11	09-May-2016	christos	s/protected/libedit_private/g
1.10	19-Apr-2016	christos	From Ingo Schwarze: - Put the data type el_rfunc_t into the public header <histedit.h>. - Make el_read in struct editline an opaque pointer rather than an embedded struct. - Do not include "read.h" everywhere, but only in the two files needing access to el_read, read.c and el.c. - To functions that don't need more, pass the struct el_read_t * rather than the full EditLine *. - Of course, that means that read_init() can now fail from memory exhaustion, but it's easy to clean up after that.
1.9	24-Feb-2016	christos	Make the read_char function always take a wchar_t * argument (Ingo Schwarze)
1.8	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.7	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.6	29-Apr-2008	martin	Convert to new 2 clause license
1.5	21-Aug-2006	christos	branches: 1.5.16; Change to a 3 clause copyright after permission of the holders.
1.4	27-Feb-2004	christos	Better fix for rl_prep_terminal() from Gerry Swislow.
1.3	21-Feb-2004	christos	Separate tty separation from the prompt printing function. From Gerry Swislow.
1.2	26-Sep-2003	christos	Implement enough of readline's 4.0 async mode to make gdb happy. This is not complete yet, but it seems to work... This required to introduce an unbuffered mode to el_gets(), but that was a minor change.
1.1	27-Sep-2001	christos	PR/14067: Anthony Mallet: Provide a programmatic way to set the read_char function via a new el_set() operation. Thanks, nicely done :-)
1.5.16.1	18-May-2008	yamt	sync with head.
1.183	14-Jun-2025	christos	Change kill(0, signo) -> raise(signo) so that we only signal the current process not the whole process group. Pointed out by geoff thomas. Related issues: https://www.postgresql.org/message-id/271520.1713052173%40sss.pgh.pa.us https://github.com/astral-sh/python-build-standalone/pull/652\ #issuecomment-2972762033 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=159226
1.182	26-Mar-2024	christos	branches: 1.182.2; fix insert key (Xose Vazquez Perez)
1.181	25-Apr-2023	christos	pass lint.
1.180	24-Apr-2023	christos	PR/57376: Jorge Giner: readline file completion does not quote; do the same.
1.179	21-Apr-2023	christos	Align types with readline-8.2 (wiz@)
1.178	02-Dec-2022	christos	PR/57095: Yilei Yang: Change readline's replace_history_entry to not make a copy of the string to replace since H_REPLACE already makes a copy (fixes memory leak)
1.177	30-Oct-2022	christos	improvements in malloc/free handling.
1.176	21-Sep-2022	christos	PR/57016: Ricky Zhou: declare lastidx
1.175	20-Sep-2022	christos	PR/57016: Ricky Zhou: Revert to trimming the last newline instead of the first one so that multi-line commands work again.
1.174	08-Apr-2022	christos	PR/56778: Detlev Casanova: Missing rl_initialize call in rl_copy_text
1.173	19-Feb-2022	christos	PR/56695: Walter Lozano: Correct declaration of hook functions.
1.172	08-Feb-2022	christos	PR/56693: Walter Lozano: Add support for rl_delete_text and rl_set_key
1.171	31-Jan-2022	christos	PR/56622: Walter Lozano: Improve readline compatibility by adding rl_readline_state support.
1.170	29-Jan-2022	christos	Add more refreshes from Walter Lozano. The readline example in http://www.mcld.co.uk/blog/2009/simple-gnu-readline-callback-style-example.html still does not work, but it is better.
1.169	11-Jan-2022	christos	PR/56618: Walter Lozano: Improve libedit compatibility with readline by implementing: rl_copy_text, rl_erase_empty_line, rl_message, rl_on_new_line, rl_replace_line, rl_restore_prompt, rl_save_prompt
1.168	10-Sep-2021	rillig	libedit: fix indentation No change to the resulting object files.
1.167	10-Sep-2021	christos	rl_startup_hook should be called each time. From Carlos Henrique Lima Melara
1.166	09-Sep-2021	christos	fix memory issues found by fuzzing (double frees and buffer overflows)
1.165	03-Sep-2021	christos	Try to refactor this in order to correct some of the memory issues reported by Christian Holler.
1.164	21-Aug-2021	christos	PR/56370: mirabilos: libedit change from 2017 kills history in gdb
1.163	21-Aug-2021	christos	Add more api to make gdb-11 happy (but not gdbtui as usual)
1.162	15-Aug-2021	rillig	libedit: simplify calls to macro ADD_STRING The lint comments CONSTCOND and LINTED were not necessary. It is simpler to just specify what to free. GCC optimizes free(NULL) to be a no-op. No functional change.
1.161	15-Aug-2021	rillig	readline: fix lint warning about effective unconst cast Calling strchr to avoid the syntactical unconst cast is not necessary here. A simple pointer assignment is enough. No functional change.
1.160	15-Aug-2021	christos	Add a LINTED comment... Why doesn't NOTREACHED work?
1.159	09-Oct-2019	christos	add +1 to strlcpy's (Patrick Welche)
1.158	08-Oct-2019	christos	Change strncpy to either memcpy (when we know the len), or strlcpy (when we used to NUL terminate explicitly.
1.157	21-Aug-2019	christos	Increment offset when adding an element to history to keep it aligned with the last element entered (Sandy Li Changqing)
1.156	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.155	07-Jun-2019	christos	PR/54281: Jonathan Perkins: NUL terminate rl_line_buffer on modification to avoid completion leak.
1.154	07-Jun-2019	christos	PR/54280: rl_completer_quote_characters should be const for readline compat
1.153	07-Jun-2019	christos	PR/54279: Jonathan Perkins: Ignore adjacent start/end prompt ignore.
1.152	26-Apr-2019	christos	Follow the man page for EL_GETTC and not require a NULL terminated argument list: https://reviews.llvm.org/D61191
1.151	15-Feb-2019	christos	PR/53983: Jonathan Perkins: Fix types for readline compatibility
1.150	14-Feb-2019	christos	PR/53981: Jonathan Perkins: history_list should null-terminate
1.149	10-Jan-2019	christos	PR/53856: F. Aragon: editline/libedit not prompting colors in readline mode
1.148	02-Dec-2018	christos	Add a couple more readline compat functions.
1.147	09-Jun-2018	christos	branches: 1.147.2; Provide more compatibility with readline headers; now python-3.6.5 works when changing 'readline' -> 'edit' in setup.py. Revert previous conditional setting of unbuffered.
1.146	01-Jan-2018	christos	branches: 1.146.2; Only FLUSH if we are ending libedit; DRAIN if we suspend for readline. This allows pasting multiline buffers (Gerry Swislow)
1.145	08-Dec-2017	christos	For applications that don't issue their own prompt (like python) don't set unbuffered unless they've already printed the prompt. This avoids printing the prompt before the application has a chance to process the input line. From sjg@
1.144	17-Sep-2017	kre	PR lib/52547 - read_history (readline.c) should now sets history_length. Patch from Yen Chi Hsuan in the PR, extracted from Apple's version of readline.c, then modified by me to be consistent about what the return value really is.
1.143	05-Sep-2017	christos	For readline emulation, don't reset the tty to "sane" (cooked) mode if we did not start this way. Also set and reset the tty on entry and exit from readline() since this is what readline does.
1.142	01-Sep-2017	christos	PR/51517: Jay West: Tty settings not restored on exit PR/51518: Jay West: prompt is interleaved with client output Both these issues are caused by rl_restore_handler not DTRT; fix it so that it kills the internal libedit state completely. This is inefficient, but it works. Also fix: 1. add append_history()/H_NSAVE_FP 2. call the rl_startup_hook before printing the first prompt as documented. callint it from rl_initialize breaks python, because the callback ends up being invoked before the readline module is installed, and we end up dereferencing a NULL pointer. 3. add el_resize_terminal. With those changes, s/lreadline/ledit/g in python works.
1.141	21-Apr-2017	abhinav	When doing filename autocompletion, append a trailing slash at the end of directory names. We already do this when there is only one completion option but in case of of multiple completion options, it wasn't being done. ok christos@
1.140	09-Jan-2017	christos	branches: 1.140.2; Make sure we take into account history_base when computing negative history offsets. (Gerry Swinslow)
1.139	28-Oct-2016	christos	pass the stream to the getc function
1.138	01-Sep-2016	mbalmer	fix typo
1.137	24-Aug-2016	christos	more compatible with readline history functions.
1.136	02-Jun-2016	christos	branches: 1.136.2; Fix previous to better match readline behavior (Ingo Schwarze)
1.135	02-Jun-2016	christos	From Ingo Schwarze: In libedit, the only way how H_ENTER can fail is memory exhaustion, too, and of course it is handled gracefully, returning -1 from history(). So of course, we will continue to handle it gracefully in add_history() as well, but we are free to decide what to do with the library state in this case because GNU just dies... I think the most reasonable course of action is to simply not change the library state in any way when add_history() fails due to memory exhaustion, but just return. If H_ENTER does not fail, we know that the history now contains at least one entry, so there is no need any longer to check the H_GETSIZE return value. And we can of course always set current_history_valid. While testing these changes, i noticed three problems so closely related that i'd like to fix them in the same diff. 1. libedit has the wrong prototype for add_history(). GNU readline-6.3 defines it as void add_history(const char ). Of course, that is very stupid - no way to report problems to the caller! But the whole point of a compatibility mode is being compatible, so we should ultimately change this. Of course, changing the prototype of a public symbol requires a libedit major bump. I don't want to do that casually. Rather, i will take a note and change the prototype the next time we need a libedit major bump for more important reasons. For now, let's just always return 0. 2. While implicitely* pushing an old entry off the history increments history_base in GNU readline, testing reveals that explicitly deleting one does not. Again, this is not documented, but it applies to both remove_history() and stifle_history(). So delete history_base manipulation from stifle_history(), which also allows to simplify the code and delete two automatic variables. 3. GNU readline add_history(NULL) crashes with a segfault. There is nothing wrong with having a public interface behave that way. Many standard interfaces do, including strlen(3). Such crashes can even be useful to catch buggy application programs. In libedit/readline.c rev. 1.104, Christos made add_history() silently ignore this coding error, according to the commit message to hide a bug in nslookup(1). That change was never merged to OpenBSD. I strongly disagree with this change. If nslookup(1) is still broken, that program needs to be fixed instead. In any case, delete the bogus check; hiding bugs is dangerous.
1.134	31-May-2016	christos	remove the right history entry (Ingo Schwarze)
1.133	13-May-2016	christos	From Bastian Maerkisch, via Igno Schwarze: Even though section "2.3.3 Information About the History List" of the history(3) info(1) manual only says -- Function: int where_history (void) Returns the offset of the current history element. which maybe isn't completely clear, a plausible implementation is that the offset returned is the same offset that can be used for history_set_pos(), i.e. that it is 0 for the oldest entry and increases with time, and that's how the GNU implementation behaves indeed. The libedit implementation, on the other hand, returns 1 for the newest entry and increases going back in time.
1.132	09-May-2016	christos	GNU readline(3) regards history chronologically, that is, from the perspective of the dawn of time, so "next" means "newer" and "previous" means "older". Libedit, by contrast, uses reverse chronology and regards history from the perspective of the present, such that "next" means "longer ago" and "previous" means "not so long ago". The following patch fixes previous_history() and next_history() as proposed by Bastian Maerkisch. But there is a related problem demonstrated by Bastian's regression tests that his patch did not fix: next_history() can advance not only to the newest entry, but beyond it, which core libedit cannot do. So that feature must be implemented locally in readline.c. With that, the last of Bastians tests is fixed, test_movement_direction(). This patch also improves libedit documentation to more clearly state what "previous" and "next" mean. GNU readline documentation is just as unclear, but we can't easily fix that since libedit doesn't include its own readline.3 manual. (Ingo Schwarze)
1.131	09-May-2016	christos	The libedit implementation of history_get() also differs from the GNU implementation: libedit goes to the entry with the given number stored in the HistEvent structure, while GNU subtracts history_base, then advances that many entries from the oldest one. If entries were removed in between, GNU advances further than libedit. The call sequence H_CURR, H_DELDATA, H_CURR, H_NEXT_EVDATA looks weird, as if part of that must somehow be redundant. But actually, the user interface is so counter-intuitive that every single step is really required. - The first H_CURR is needed to be able to go back after an error. - The H_DELDATA is needed to move the cursor. Even though it takes a pointer to ev, that structure is not filled in when the call succeeds. H_DELDATA only moves the cursor, it doesn't tell us the new event number. - Consequently, the second H_CURR is required to get ev.num filled in. But it doesn't return the data because ev has no field for that. - So even though the cursor is already positioned correctly, H_NEXT_EVDATA is needed as the final step merely to get the data. (Ingo Schwarze)
1.130	08-May-2016	christos	In stiffle_history(), trim excessive entries from the history and advance history_base like the GNU implementation does. (from Bastian Maerkisch)
1.129	06-May-2016	christos	fix logic (Ingo Schwarze)
1.128	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.127	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.126	24-Feb-2016	christos	Make the read_char function always take a wchar_t * argument (Ingo Schwarze)
1.125	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.124	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.123	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.122	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.121	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.120	15-Feb-2016	christos	attribute unused
1.119	15-Feb-2016	christos	OpenBSD readline.c rev. 1.14 2015/02/06 23:21:58 millert use SIZE_MAX
1.118	15-Feb-2016	christos	OpenBSD readline.c rev. 1.13 2015/01/13 08:33:12 reyk rl_set_keyboard_input_timeout() for readline 4.2 compat
1.117	02-Jun-2015	christos	Adjust API to a more modern readline (Ryo Onodera)
1.116	26-May-2015	christos	- fix types of rl_completion_entry_function and rl_add_defun - call update pos before completion to refresh the screen From Thomas Eriksson
1.115	01-Apr-2015	christos	Fix overlapping strcpy (Gerry Swislow)
1.114	24-Mar-2015	christos	set some readline compatibility default key settings. https://bugzilla.redhat.com/attachment.cgi?id=1001895
1.113	18-Oct-2014	snj	src is too big these days to tolerate superfluous apostrophes. It's "its", people!
1.112	15-Aug-2014	christos	Fix typo in comment (Tobias Stoeckmann)
1.111	06-Jul-2014	christos	branches: 1.111.2; PR/48957: Federico G. Schwindt: Restore commented out code that broke rl_callback_handler.
1.110	21-Jan-2014	christos	branches: 1.110.2; ... if called prior to using_history(). This needed to be worked around in PHP: http://git.php.net/?p=php-src.git;a=commitdiff;h=31d67bd3 Bug: https://bugzilla.redhat.com/show_bug.cgi?id=1055409
1.109	28-Aug-2013	christos	get rid of PATH_MAX.
1.108	28-May-2013	christos	expose rl_catch_signals and explain what we are doing.
1.107	13-Jan-2013	christos	explicitly pass (void *)0 instead of NULL.
1.106	12-Oct-2012	christos	Add trailing NULL's to the varargs functions as required. (John Spencer)
1.105	12-Jul-2012	christos	branches: 1.105.2; PR/46678: Ian Wienand: Add stub implementation for rl_free_line_state()
1.104	05-Jun-2012	christos	don't crash if add_history is called from an empty line. Called from nslookup in new bind. XXX: pullup to 6
1.103	15-May-2012	christos	define the new variable
1.102	15-May-2012	christos	Add rl_completion_word_break_hook from: http://gitweb.dragonflybsd.org/dragonfly.git/commitdiff/\ 5ec6a45fa762b8cbd0305ca06acb8780335a486a
1.101	21-Mar-2012	matt	Use C89 functions definitions. Remove use of __P
1.100	18-Nov-2011	christos	branches: 1.100.2; Add missing *
1.99	16-Aug-2011	christos	branches: 1.99.2; re-enable -Wconversion
1.98	29-Jul-2011	christos	pass -Wconversion
1.97	29-Jul-2011	christos	- fix unused params - unconditionalize vis.h
1.96	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.95	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.94	28-Jul-2011	christos	handle systems without getpwent_r
1.93	28-Jul-2011	christos	use vis.h from .
1.92	16-Sep-2010	christos	unbreak readline history.
1.91	28-Aug-2010	christos	setup a callback to be invoked on resize buffers so that readline can reset rl_line_buffer which unfortunately some applications use it directly.
1.90	04-Aug-2010	christos	provide rl_on_newline
1.89	15-Apr-2010	christos	From Jess Thrysoee - Fix wint_t to Int confusion
1.88	03-Jan-2010	christos	rename historyw -> history_w for consistency. add wide tst code and make it the default.
1.87	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.86	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.85	07-Sep-2009	christos	apply apple patches from: http://opensource.apple.com/source/libedit/libedit-11/patches/
1.84	22-Jul-2009	christos	Fix memory leaks in error paths. From Michael Cook mcook at bbn dot com
1.83	08-Apr-2009	christos	Fix off by one error reported by: Caleb Welton cwelton at greenplum dot com
1.82	31-Mar-2009	christos	implement RL_PROMPT_{START,END}_IGNORE
1.81	21-Feb-2009	christos	more size_t stuff.
1.80	15-Feb-2009	christos	pass lint on _LP64.
1.79	12-Feb-2009	sketch	More fixes for existing portability stuff.
1.78	05-Feb-2009	christos	branches: 1.78.2; add rl_set_prompt
1.77	18-Jan-2009	lukem	fix -Wsign-compare issues
1.76	11-Jan-2009	christos	- insert a space after the recognized string if it was an exact match - initialize properly the string used for completion. From Alex Bligh alex at alex dot org dot uk - Make char constants consistent
1.75	29-Apr-2008	martin	Convert to new 2 clause license
1.74	05-Apr-2008	christos	branches: 1.74.2; add EL_REFRESH for the benefit of readline
1.73	04-Apr-2008	christos	Add rl_forced_update_display() from Gerry Swislow
1.72	12-Aug-2007	christos	patches from Axel Thimm
1.71	27-May-2007	christos	Add rl_completion_matches, fix remove_history
1.70	24-Nov-2006	christos	- Add more readline functions, enough for gdb-6.5 - Make el_get varyadic, and implement EL_GETTC. - XXX: the EL_SETTC api will change in the future.
1.69	21-Aug-2006	christos	Change to a 3 clause copyright after permission of the holders.
1.68	23-Jul-2006	christos	PR/34062: Tanaka Akira: rl_deprep_term_function is NULL in libedit. Default to rl_deprep_terminal as suggested; do the same for rl_prep_term_function
1.67	25-Mar-2006	rtr	remove if/free block checking known condition coverity 2762 / run 11
1.66	21-Mar-2006	christos	Coverity CID 2743: Not really a memory leak, but make it obvious that we always free tmp.
1.65	18-Mar-2006	christos	Lint comment.
1.64	18-Mar-2006	christos	Coverity CID 1666: Plug memory leak.
1.63	18-Mar-2006	christos	Coverity CID 1667: Plug memory leak
1.62	18-Mar-2006	christos	Coverity CID 1662: Memory leak.
1.61	06-Mar-2006	christos	use the tty chars for reprint and eof instead of hard-coded ^R and ^D
1.60	13-Feb-2006	christos	PR/32817: Magnus Svensson: write_history and read_history returncode is not readline compatible.
1.59	12-Feb-2006	christos	Partial rl_getc_function support from Jess Thrysoee.
1.58	14-Jul-2005	christos	PR/30747: David N. Williams: libedit is missing remove_history() Added, please test.
1.57	11-Jun-2005	christos	PR/30500: Paul Shupak: Inconsistent definition of tilde_expand(). Provide a layer of indirection between the readline compatibility functions and our internal implementation, so that we have the freedom to change the function signature.
1.56	10-Jun-2005	christos	Bug reported from Martin Dietze: The place to change the completion_append_character is usually somewhere in the `rl_completion_entry_function' callback which is where one usually can distinguish between file- or dir-like entries to append a slash for dirs etc. This does no longer work since `fn_complete()' takes the `append_character' as argument before the callback is executed, so that changes to the variable `rl_completion_append_character' have in fact no effect for the current completion. Fix by adding a function that returns the rl_completion_append_character, when it gets passed in a filename in readline emulation.
1.55	27-May-2005	agc	Use the correct type for the stored callback function
1.54	27-May-2005	agc	Sync the alternative readline interface with reality: + the rl_callback_handler_install takes a pointer to a void function which has one char * argument (it's called that way in the readline emulation source, otherwise there's no way to pass the line buffer to the function which processes the line when EOL is encountered) + provide a prototype for that function signature and use it Makes the callback readline interface work now.
1.53	07-May-2005	dsl	Separate out the filename completion functions from the readline() code. Pass in loads of parameters instead of relying on shed-loads of global variables to modify the behaviour. The filename completion code can now be enabled by code that uses el_gets(). (eg /bin/sh)
1.52	19-Apr-2005	christos	check for pwd != NULL, fix a missed getpwnam.
1.51	12-Apr-2005	christos	PR/29958: Peter Bex: add rl_variable_bind and rl_attempted_completion_over
1.50	02-Apr-2005	christos	use getpwent_r
1.49	10-Mar-2005	christos	branches: 1.49.2; Always update the position variables before the map function is called. From Rob Rodgers, thanks!
1.48	09-Mar-2005	christos	set UNBUFFERED again after you do the line callback so that the new line gets refreshed.
1.47	08-Sep-2004	christos	cut out the middle-man and use el_insertstr() directly.
1.46	27-Feb-2004	christos	Better fix for rl_prep_terminal() from Gerry Swislow.
1.45	17-Jan-2004	christos	portability fixes.
1.44	05-Dec-2003	lukem	Tokenization function enhancements: * Make tok_init(), tok_end(), tok_reset(), tok_line() and tok_str() publically available in <histedit.h> * Documented the public functions in editline(3) * Renamed tok_line() -> tok_str() * Added new tok_line() which takes a "const LineInfo " instead of "const char " (the former has "cursor" information), and optionally return the argv index ("int cursorc") and offset within that index ("int cursorv"). This means that completion routines can use the tokenization code to crack the line and easily find which word the cursor is at. (mmm, context sensitive completion :) * Fixed TEST/test.c when using "continuation" lines (unmatched quote or \ at EOL), and added some more DEBUG messages including highlighting where the cursor is (with a `_').
1.43	03-Nov-2003	christos	Fix uninitialized variable.
1.42	02-Nov-2003	christos	From: Gerry Swislow gerry at certif dot com 1) File name completion should list the files in the current directory if no text is entered. The previous version wouldn't list anything if the text to complete was empty. 2) When listing directories, the entries "." and ".." shouldn't be shown. 3) The filename completion should be used if the user's rl_attempted_completion_function doesn't return any matches. The previous version didn't do that.
1.41	01-Nov-2003	christos	initialize ptr.
1.40	27-Oct-2003	christos	Make readline csh-like history work. From Gerry Swislow <gerry at certif dot com>.
1.39	19-Oct-2003	christos	add rl_catch_signals
1.38	16-Oct-2003	christos	More libedit readline emulation functions from: Gerry Swislow <gerry at certif dot com>
1.37	15-Oct-2003	christos	don't limit ^c to alpha c, and add VIS_NOSLASH so that vis(3) does not produce \^c
1.36	09-Oct-2003	christos	PR/23107: Nathan Williams: ^D as the first char on the command line does not DTRT in readline compatibility mode
1.35	26-Sep-2003	christos	Implement enough of readline's 4.0 async mode to make gdb happy. This is not complete yet, but it seems to work... This required to introduce an unbuffered mode to el_gets(), but that was a minor change.
1.34	14-Sep-2003	christos	Match the stupid function pointer declarations with actual readline's 4.0. This is gross.
1.33	14-Sep-2003	christos	make this compile.
1.32	14-Sep-2003	christos	- provide enough hooks to compile gdb-5.3 - fix el_get(e, EL_TERMINAL, (char **))
1.31	19-Jun-2003	christos	From michael@moria.de: Fix realloc case where we could be running out of space if too many matches.
1.30	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.29	29-Mar-2003	wiz	Consistently spell occurrence with two rs.
1.28	10-Mar-2003	christos	null is not 0
1.27	10-Mar-2003	christos	include alloca.h for systems that need it.
1.26	21-Jan-2003	christos	fix directory descriptor leak [from michael at moria dot de]
1.25	27-Oct-2002	christos	remove unused variables.
1.24	27-Oct-2002	christos	vi mode and memory fixes from david laight.
1.23	27-Oct-2002	christos	don't crash in memory shortage conditions.
1.22	09-Apr-2002	thorpej	_rl_qsort_string_compare(): Fix casts. This makes gcc 3.2 happy, and also allows both LINTED comments to be removed.
1.21	18-Mar-2002	christos	Add a couple linted comment and enable WARNS=3
1.20	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.19	10-Jan-2001	jdolecek	fix partial completion - we failed to refresh screen in that case
1.18	05-Jan-2001	christos	depoison the pure editline code from readline compatibility hacks.
1.17	05-Jan-2001	jdolecek	Standard location of readline headers is /usr/include/readline/, so install them there. readline.h of libedit had to move to subdirectory 'readline', due to the way BSD makefiles work; this is better than potentially fragile Makefile hacks
1.16	04-Jan-2001	christos	fix lint problems.
1.15	01-Jan-2001	jdolecek	Add support for rl_completion_append_character and rl_special_prefixes. This addresses lib/10513 by Giles Lean. Tested with PostgreSQL 7.0.2 psql.
1.14	01-Jan-2001	jdolecek	history_tokenize(): fix one off-by-one bug rl_complete_internal(): only replace the completed string with common part of possible matches if there is a possible completion
1.13	31-Dec-2000	jdolecek	rl_display_match_list(): * pad entries shorter than 'max' by spaces correctly * fix off-by-one error which caused extra newline to be printed if the list fit exactly to a screen * fix typo in _rl_qsort_string_compare, which caused the list to not be sorted after all
1.12	23-Dec-2000	jdolecek	completion_matches(): fix a off-by-one bug, fix variable name typo implement displaying of possible completions, add hook to display the list on second rl_complete() invocation in row (typically, double <TAB>) This addresses the completion part of lib/11581 by Richard Earnshaw.
1.11	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.10	10-Mar-2000	jdolecek	Fix _rl_compat_sub() to really honour 'globally' flag instead of making the substitution always globally - affects _history_expand_command() and in turn history_expand() All praise lint :)
1.9	10-Mar-2000	jdolecek	readline(): don't dereference NULL pointer if prompt is NULL history_tokenize(): avoid stepping too far if backslash is last character on the passed string update some comments This makes gdb \ escaping work reliably and fixes lib/9511 by Assar Westerlund.
1.8	03-Jul-1999	lukem	don't try to free() something allocated with alloca() in rl_complete_internal(). (noticed while using completion in gdb and getting inundated with warnings from the new free())
1.7	02-Jul-1999	simonb	More trailing white space.
1.6	12-Jun-1999	christos	Make this compile under linux
1.5	12-Dec-1998	christos	many problems; variables hidden by others, size_t <-> confusion
1.4	20-May-1998	christos	Adjust to the libedit api changes.
1.3	12-Nov-1997	thorpej	el_gets() takes an int , not a size_t .
1.2	23-Oct-1997	christos	Const de-poisoning :-( Unfortunately the default gnu readline does not have full prototypes... With those changes, and a single line change in gdb/top.c, gdb links with -ledit
1.1	23-Oct-1997	christos	PR/4301: Jaromir Dolecek. Add gnu-readline wrapper for editline.
1.49.2.4	12-Jul-2005	tron	Pull up revision 1.52 (requested by lukem in ticket #542): check for pwd != NULL, fix a missed getpwnam.
1.49.2.3	12-Jul-2005	tron	Pull up revision 1.50 (requested by lukem in ticket #542): use getpwent_r
1.49.2.2	28-May-2005	tron	Pull up revision 1.55 (requested by agc in ticket #353): Use the correct type for the stored callback function
1.49.2.1	28-May-2005	tron	Pull up revision 1.54 (requested by agc in ticket #353): Sync the alternative readline interface with reality: + the rl_callback_handler_install takes a pointer to a void function which has one char * argument (it's called that way in the readline emulation source, otherwise there's no way to pass the line buffer to the function which processes the line when EOL is encountered) + provide a prototype for that function signature and use it Makes the callback readline interface work now.
1.74.2.1	18-May-2008	yamt	sync with head.
1.78.2.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.99.2.5	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.99.2.4	23-Jan-2013	yamt	sync with head
1.99.2.3	30-Oct-2012	yamt	sync with head
1.99.2.2	23-May-2012	yamt	sync with head.
1.99.2.1	17-Apr-2012	yamt	sync with head
1.100.2.1	05-Jun-2012	bouyer	Pull up following revision(s) (requested by christos in ticket #309): lib/libedit/readline.c: revision 1.104 don't crash if add_history is called from an empty line. Called from nslookup in new bind. XXX: pullup to 6
1.105.2.4	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.105.2.3	23-Jun-2013	tls	resync from head
1.105.2.2	25-Feb-2013	tls	resync with head
1.105.2.1	20-Nov-2012	tls	Resync to 2012-11-19 00:00:00 UTC
1.110.2.1	10-Aug-2014	tls	Rebase.
1.111.2.1	13-May-2015	martin	Sync lib/libedit with head, requested by christos in #753: lib/libedit/Makefile 1.53 lib/libedit/chartype.h 1.13 lib/libedit/editline.3 1.83-1.84 lib/libedit/editrc.5 1.28-1.29 lib/libedit/eln.c 1.18 lib/libedit/filecomplete.c 1.33-1.34 lib/libedit/readline.c 1.112-1.115 Man page improvements, fix overlapping strcpy, improve readline compatibility, clang build fix.
1.136.2.3	26-Apr-2017	pgoyette	Sync with HEAD
1.136.2.2	20-Mar-2017	pgoyette	Sync with HEAD
1.136.2.1	04-Nov-2016	pgoyette	Sync with HEAD
1.140.2.1	21-Apr-2017	bouyer	Sync with HEAD
1.146.2.3	18-Jan-2019	pgoyette	Synch with HEAD
1.146.2.2	26-Dec-2018	pgoyette	Sync with HEAD, resolve a few conflicts
1.146.2.1	25-Jun-2018	pgoyette	Sync with HEAD
1.147.2.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.147.2.1	10-Jun-2019	christos	Sync with HEAD
1.182.2.1	02-Aug-2025	perseant	Sync with HEAD
1.6	05-Jan-2001	jdolecek	Standard location of readline headers is /usr/include/readline/, so install them there. readline.h of libedit had to move to subdirectory 'readline', due to the way BSD makefiles work; this is better than potentially fragile Makefile hacks
1.5	01-Jan-2001	jdolecek	Add support for rl_completion_append_character and rl_special_prefixes. This addresses lib/10513 by Giles Lean. Tested with PostgreSQL 7.0.2 psql.
1.4	23-Dec-2000	jdolecek	completion_matches(): fix a off-by-one bug, fix variable name typo implement displaying of possible completions, add hook to display the list on second rl_complete() invocation in row (typically, double <TAB>) This addresses the completion part of lib/11581 by Richard Earnshaw.
1.3	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.2	23-Oct-1997	christos	Const de-poisoning :-( Unfortunately the default gnu readline does not have full prototypes... With those changes, and a single line change in gdb/top.c, gdb links with -ledit
1.1	23-Oct-1997	christos	PR/4301: Jaromir Dolecek. Add gnu-readline wrapper for editline.
1.60	05-Dec-2024	christos	Don't eat 0 width characters, print them.
1.59	30-Jun-2024	christos	Handle the case where the cursor is on the first character. set -o vi x ESC ~ (Robert Morris) https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=279545
1.58	09-Sep-2021	christos	branches: 1.58.4; Add casts to appease conversions between wchar_t and wint_t
1.57	30-Mar-2020	ryo	fix build error with SDEBUG, MAP_DEBUG, DEBUG_REFRESH
1.56	04-Jan-2019	uwe	Fix mis-applied change in previous. Don't increment r_oldcv twice. PR lib/53803
1.55	23-Oct-2018	christos	Apply revisions 1.21, 1.22 from OpenBSD: In re_fastputc(), set lastline to the new line, not the previous line so it gets initialized properly. Fixes a crash in bc with MALLOC_OPTIONS=UJ. OK deraadt@, committing on behalf of yasuoka@ Initialize "old" screen buffer lines before use; otherwise, they would never get NUL-terminated and cause read buffer overruns. This fixes for example segfaults in sftp(1) that could be triggered by typing in an extremely long string (more than one line - the longer, the likelier to crash), then hitting backspace once. Problem reported and patch OK'ed by sthen@. XXX: pullup-8
1.54	30-Jun-2017	kre	branches: 1.54.4; 1.54.6; Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale).
1.53	27-Jun-2017	christos	fix comment
1.52	27-Jun-2017	christos	- add literal sequence handling.
1.51	09-May-2016	christos	branches: 1.51.8; s/protected/libedit_private/g
1.50	02-May-2016	christos	fix typos from Pedro Giffuni @FreeBSD
1.49	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.48	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.47	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.46	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.45	02-Mar-2016	christos	PR/50880: David Binderman: Remove redundant code. While here, fix all debugging formats.
1.44	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.43	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.42	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.41	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.40	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.39	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.38	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.37	29-Jul-2011	christos	pass -Wconversion
1.36	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.35	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.34	28-Dec-2009	christos	reduce diff with tcsh
1.33	28-Dec-2009	christos	Fix bug where tab completion on the second or > line that caused listing ended up corrupting the display by an extra space in the beginning. Reported by Mac Chan.
1.32	17-Jul-2009	christos	Simplify the code. No functional change.
1.31	19-May-2009	christos	always scroll when we advance past bottom. From Caleb Welton cwelton at greenplum dot com
1.30	31-Mar-2009	christos	Implement literal prompt sequences. Now someone can implement RL_PROMPT_START_LITERAL/RL_PROMPT_END_LITERAL :-)
1.29	15-Feb-2009	christos	pass lint on _LP64.
1.28	10-Sep-2008	christos	branches: 1.28.6; Allow a single process to control multiple ttys (for pthreads using _REENTRANT) using multiple EditLine objects. Mostly from Preston A. Elder.
1.27	09-Nov-2005	christos	branches: 1.27.20; Refresh bug reported by Julien Torres: going from: activate -verbose to: reset -activation results in: reset -activationverbose" instead of: reset -activation This is because we choose to insert "reset -" before the current line, and the delete "e -" and insert "ion" in the appropriate place. The cleareol code did not handle this case properly; we now cleareol to the maximum number of characters of the first difference, the second difference and the difference in line length.
1.26	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.25	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.24	10-Mar-2003	christos	s/u_int32_t/unsigned int/
1.23	15-Nov-2002	christos	de-lint
1.22	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.21	12-Nov-2002	thorpej	Fix signed/unsigned comparison warnings.
1.20	27-Oct-2002	christos	vi mode and memory fixes from david laight.
1.19	27-Oct-2002	christos	PR/17954: SAITOH Masanobu: since we have opost and onlcr set, we don't need to output \r to go to the end of line; \n is enough. From David Laight.
1.18	18-Mar-2002	christos	branches: 1.18.2; - constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.17	13-Apr-2001	lukem	rename 3 arg ELRE_DEBUG to ELRE_ASSERT, add 2 arg ELRE_DEBUG, and change all occurences of ELRE_DEBUG(foo,bar,) -> ELRE_DEBUG(foo,bar). some compilers (e.g, gcc on darwin) bitch about the former (`not enough args').
1.16	10-Jan-2001	jdolecek	Enlarge editline buffers as needed to support arbitrary length lines. This also addresses lib/9712 by Phil Nelson.
1.15	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.14	13-Mar-2000	soren	branches: 1.14.4; Fix doubled 'the's.
1.13	19-Feb-2000	mycroft	Fix refresh glitches when using auto-margin.
1.12	20-Jan-2000	christos	Add support for automatic and magic margins (from tcsh) This makes the rightmost column usable on all programs that use editline.
1.11	13-Nov-1999	lukem	instead of using a private coord_t global variable to store the size of the rprompt, use the previously unused coord_t el->el_rprompt.p_pos
1.10	12-Nov-1999	lukem	- implement printing a right-side prompt. code derived from similar work I wrote for tcsh(1) three years ago. - implement EL_RPROMPT, which allows a setting/getting of a function which returns a string to be used as the right-side prompt. - improve HISTORY and AUTHORS sections in editline(3). - bump shlib minor version for EL_RPROMPT. XXX: due to an implementation issue, the rprompt has a 1 space gap before the edge of the logical screen. editline's logical screen is 1 space less than the full screen width, so there's a 2 space gap between the rprompt and the right end of the physical screen. i'm not concerned about this.
1.9	15-Oct-1999	jdolecek	don't assume locales are not working - it may not be the case re_refresh(): cast the character passed to re_addc() to unsigned char, so we don't end up calling isprint() with negative value when chars are signed and character value is >= 128
1.8	02-Jul-1999	simonb	branches: 1.8.2; More trailing white space.
1.7	12-Jun-1999	christos	Make this compile under linux
1.6	05-Feb-1999	christos	delint.
1.5	12-Dec-1998	christos	delint
1.4	20-May-1998	christos	cast is*() arg to unsigned char
1.3	06-Jul-1997	christos	Fix compiler warnings.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.8.2.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.14.4.1	15-Mar-2003	he	Pull up revision 1.19 (via patch, requested by msaitoh in ticket #14): Since we have opost and onlcr set, we don't need to output \r; \n is sufficient. Fixes PR#17954.
1.18.2.1	05-Jan-2003	jmc	Pull up revisions 1.18-1.19 (requested by masanobu in ticket #1048) PR/17954: SAITOH Masanobu: since we have opost and onlcr set, we don't need to output \r to go to the end of line; \n is enough. From David Laight.
1.27.20.1	24-Sep-2008	wrstuden	Merge in changes between wrstuden-revivesa-base-2 and wrstuden-revivesa-base-3.
1.28.6.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.51.8.1	23-Jul-2017	snj	Pull up following revision(s) (requested by kre in ticket #102): lib/libedit/Makefile: 1.64-1.65 lib/libedit/editline.3: 1.94-1.96 lib/libedit/editrc.5: 1.33 lib/libedit/el.c: 1.93-1.94 lib/libedit/el.h: 1.42 lib/libedit/literal.c: 1.1-1.3 lib/libedit/literal.h: 1.1-1.2 lib/libedit/prompt.c: 1.27 lib/libedit/read.c: 1.103 lib/libedit/refresh.c: 1.52-1.54 lib/libedit/refresh.h: 1.11 lib/libedit/terminal.c: 1.33 Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming. -- Include EDITRC in doc. -- mention the limitation of the literal sequence delimiter. -- - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display. -- - add literal sequence handling. -- remove unused variable -- add literal escape sequence support, patterned after the tcsh ones. -- fix comment -- Fix an obvious, but almost invisible typo (avoid some core dumps). -- Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale). -- Remove workaround for ancient HTML generation code.
1.54.6.2	08-Apr-2020	martin	Merge changes from current as of 20200406
1.54.6.1	10-Jun-2019	christos	Sync with HEAD
1.54.4.2	18-Jan-2019	pgoyette	Synch with HEAD
1.54.4.1	26-Nov-2018	pgoyette	Sync with HEAD, resolve a couple of conflicts
1.58.4.2	02-Aug-2025	perseant	Sync with HEAD
1.58.4.1	01-Jul-2024	perseant	Sync with HEAD.
1.11	27-Jun-2017	christos	- add literal sequence handling.
1.10	09-May-2016	christos	branches: 1.10.8; s/protected/libedit_private/g
1.9	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.8	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.7	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.6	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.5	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.4	10-Jan-2001	jdolecek	Enlarge editline buffers as needed to support arbitrary length lines. This also addresses lib/9712 by Phil Nelson.
1.3	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.10.8.1	23-Jul-2017	snj	Pull up following revision(s) (requested by kre in ticket #102): lib/libedit/Makefile: 1.64-1.65 lib/libedit/editline.3: 1.94-1.96 lib/libedit/editrc.5: 1.33 lib/libedit/el.c: 1.93-1.94 lib/libedit/el.h: 1.42 lib/libedit/literal.c: 1.1-1.3 lib/libedit/literal.h: 1.1-1.2 lib/libedit/prompt.c: 1.27 lib/libedit/read.c: 1.103 lib/libedit/refresh.c: 1.52-1.54 lib/libedit/refresh.h: 1.11 lib/libedit/terminal.c: 1.33 Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming. -- Include EDITRC in doc. -- mention the limitation of the literal sequence delimiter. -- - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display. -- - add literal sequence handling. -- remove unused variable -- add literal escape sequence support, patterned after the tcsh ones. -- fix comment -- Fix an obvious, but almost invisible typo (avoid some core dumps). -- Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale). -- Remove workaround for ancient HTML generation code.
1.52	30-Jun-2024	christos	Prevent reading before the line buffer (try ^R^W in emacs mode at the beginning of the line) (Robert Morris) https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=279426
1.51	30-Mar-2020	ryo	branches: 1.51.8; patbuf must be updated if the length of patbuf is greater than or equal to 0. (that is always) fix of r1.7 was incorrect.
1.50	30-Mar-2020	ryo	fix build error with SDEBUG, MAP_DEBUG, DEBUG_REFRESH
1.49	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.48	26-Feb-2018	christos	branches: 1.48.4; PR/53058: Nikhil Benesch: use correctly typed variables (wchar_t vs wint_t) as parameters.
1.47	09-May-2016	christos	s/protected/libedit_private/g
1.46	28-Apr-2016	christos	Initialize patbuf (Ingo Schwarze)
1.45	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.44	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.43	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.42	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.41	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.40	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.39	24-Feb-2016	christos	Get split el_getc and el_wgetc completely and call el_wgetc internally. Change some character constants to they wide versions. (Ingo Schwarze)
1.38	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.37	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.36	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.35	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.34	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.33	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.32	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.31	30-Jan-2016	christos	Fix misplaced parentheses (Ingo Schwarze)
1.30	04-Oct-2011	christos	fixed warnings where wint_t is unsigned.
1.29	16-Aug-2011	christos	re-enable -Wconversion
1.28	29-Jul-2011	christos	pass -Wconversion
1.27	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.26	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.25	28-Jul-2011	christos	term -> terminal
1.24	15-Apr-2010	christos	From Jess Thrysoee - Fix wint_t to Int confusion
1.23	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.22	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.21	15-Feb-2009	christos	pass lint on _LP64.
1.20	04-Nov-2004	christos	branches: 1.20.34; Make EM_DELETE_PREV_CHAR behave like ED_DELETE_PREV_CHAR in incremental search. From Gerry Swislow.
1.19	25-Oct-2003	christos	Another fix for incremental search prev.
1.18	18-Oct-2003	christos	make forward incremental search work better.
1.17	17-Oct-2003	christos	When searching backwards don't include the characters after the cursor in the search.
1.16	16-Oct-2003	christos	Fix incremental search which was badly busted.
1.15	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.14	20-Nov-2002	christos	Fix problem with previous patches that broke vi history. - c_gets() was usually returning a length, but sometimes one of the CC_xxx values (which are small +ve integers)! - fixed c_gets() by putting a ' ' under the cursor. From David Laight.
1.13	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.12	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.11	23-Jan-2001	jdolecek	sprinkle couple const
1.10	04-Jan-2001	christos	consistently check for allocation failures and return -1, if we could not get more memory.
1.9	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.8	02-Jul-1999	simonb	More trailing white space.
1.7	05-Feb-1999	christos	delint.
1.6	02-Sep-1998	christos	PR/6081: Wolfgang Helbig: search broken in vi mode. When patbuf was changed to be dynamically allocated, sizeof was not changed appropriately.
1.5	06-Jul-1997	christos	Fix compiler warnings.
1.4	23-Jan-1997	mrg	- convert unsafe strcpy(), strcat() and sprintf() to the `n' versions. - some KNF.
1.3	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.2	02-Oct-1994	jtc	Added code so that POSIX.2 regular expresion functions are used if REGEX is defined, V8 regular expresion functions are used if REGEXP is defined, and BSD regular expression functions are used if neither are defined. And defined REGEX in sys.h so that programs using libedit don't have to link with libcompat.
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	branches: 1.1.1.1.2; libedit!
1.1.1.1.2.1	07-Oct-1994	mycroft	Update from trunk.
1.20.34.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.48.4.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.48.4.1	08-Apr-2020	martin	Merge changes from current as of 20200406
1.51.8.1	01-Jul-2024	perseant	Sync with HEAD.
1.14	09-May-2016	christos	s/protected/libedit_private/g
1.13	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.12	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.11	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.10	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.9	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.8	18-Oct-2003	christos	make forward incremental search work better.
1.7	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.6	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.5	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.4	02-Jul-1999	simonb	More trailing white space.
1.3	12-Dec-1998	christos	delint
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.19	22-Jan-2013	christos	provide an el_init_fd function.
1.18	11-Jan-2009	christos	branches: 1.18.8; 1.18.14; bump shared libraries.
1.17	05-Apr-2008	christos	bump minor.
1.16	24-Nov-2006	christos	- Add more readline functions, enough for gdb-6.5 - Make el_get varyadic, and implement EL_GETTC. - XXX: the EL_SETTC api will change in the future.
1.15	05-Dec-2003	lukem	Tokenization function enhancements: * Make tok_init(), tok_end(), tok_reset(), tok_line() and tok_str() publically available in <histedit.h> * Documented the public functions in editline(3) * Renamed tok_line() -> tok_str() * Added new tok_line() which takes a "const LineInfo " instead of "const char " (the former has "cursor" information), and optionally return the argv index ("int cursorc") and offset within that index ("int cursorv"). This means that completion routines can use the tokenization code to crack the line and easily find which word the cursor is at. (mmm, context sensitive completion :) * Fixed TEST/test.c when using "continuation" lines (unmatched quote or \ at EOL), and added some more DEBUG messages including highlighting where the cursor is (with a `_').
1.14	26-Sep-2003	christos	Implement enough of readline's 4.0 async mode to make gdb happy. This is not complete yet, but it seems to work... This required to introduce an unbuffered mode to el_gets(), but that was a minor change.
1.13	14-Sep-2003	christos	- provide enough hooks to compile gdb-5.3 - fix el_get(e, EL_TERMINAL, (char **))
1.12	09-Oct-2001	christos	PR/14188: Anthony Mallet: Provide an opaque data pointer to client programs.
1.11	01-Jan-2001	jdolecek	bump libedit minor - addition of rl_special_prefixes and rl_completion_append_character
1.10	23-Dec-2000	jdolecek	bump libedit minor - added some public stuff for readline emulation
1.9	12-Nov-1999	lukem	- implement printing a right-side prompt. code derived from similar work I wrote for tcsh(1) three years ago. - implement EL_RPROMPT, which allows a setting/getting of a function which returns a string to be used as the right-side prompt. - improve HISTORY and AUTHORS sections in editline(3). - bump shlib minor version for EL_RPROMPT. XXX: due to an implementation issue, the rprompt has a 1 space gap before the edge of the logical screen. editline's logical screen is 1 space less than the full screen width, so there's a 2 space gap between the rprompt and the right end of the physical screen. i'm not concerned about this.
1.8	25-Feb-1999	abs	branches: 1.8.6; Add a note to update src/distrib/sets/lists/base/shl.*, and add a missing RCS Id.
1.7	29-Jul-1998	lukem	* add more checks for NULL pointers in passed arguments * implement el_get(EditLine , int op, void result), which does the inverse of el_set() * add EL_EDITMODE operation to el_set and el_get; if non zero editing is enabled (the default). * add "edit on \| off" editrc command, which modifies EL_EDITMODE. users can now add '*:edit off' in ~/.editrc as an advisory to disable editing. NOTE: at this time EL_EDITMODE is just an indication of the state of the 'edit' command. It's up to the application to check this after el_source() or el_parse() to determine if editing is still required.
1.6	01-Jun-1998	lukem	* implement CC_REFRESH_BEEP; as per CC_REFRESH but beep as well. this is useful in completion when a partial completion is found * remove entry in BUGS about el_parse(); that was fixed a while ago
1.5	20-May-1998	christos	Bump!
1.4	05-Jan-1998	perry	RCSID Police.
1.3	23-Oct-1997	christos	PR/4301: Jaromir Dolecek. Add gnu-readline wrapper for editline.
1.2	14-Oct-1997	christos	PR/4257: Jaromir Dolecek: history() has no generic error handling and isn't reentrant. This changes the interface of the history function, so we need a major number bump.
1.1	06-May-1994	cgd	local
1.8.6.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.18.14.1	25-Feb-2013	tls	resync with head
1.18.8.1	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.29	14-Jun-2025	christos	Change kill(0, signo) -> raise(signo) so that we only signal the current process not the whole process group. Pointed out by geoff thomas. Related issues: https://www.postgresql.org/message-id/271520.1713052173%40sss.pgh.pa.us https://github.com/astral-sh/python-build-standalone/pull/652\ #issuecomment-2972762033 https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=159226
1.28	18-Dec-2024	christos	From tkoeppe@github: Specify SA_ONSTACK when setting up a signal handler. This allows the handler to use the alternate signal stack if one is available in the handling thread, but has no effect otherwise. This change makes the signal handler respect existing choices better. Specifically, this allows signal handlers to be set when the process includes a Go runtime, since Go enforces that all signal handlers in the process use the SA_ONSTACK flag (e.g. see golang/go#20400).
1.27	03-Feb-2023	christos	branches: 1.27.2; Don't clear the handle to el "sel" and set it earlier to prevent handling a signal before the handle is set. (From: des at FreeBSD)
1.26	09-May-2016	christos	s/protected/libedit_private/g
1.25	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.24	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.23	16-Feb-2016	christos	include errno.h
1.22	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.21	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.20	16-Feb-2016	christos	one more
1.19	16-Feb-2016	christos	include explicitly errno.h since we use it.
1.18	15-Feb-2016	christos	OpenBSD sig.c rev. 1.6 2001/12/06 04:26:00 deraadt save and restore errno in signal handler
1.17	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.16	28-Jul-2011	christos	term -> terminal
1.15	19-Feb-2009	christos	reset and redraw on sigcont. From Anon Ymous.
1.14	18-Feb-2009	christos	SA_RESTART for all signals but SIGINT. From Anon Ymous.
1.13	15-Feb-2009	christos	in order for read() to return EINTR we need to use sigaction, not signal, otherwise SA_RESTART is set.
1.12	10-Sep-2008	christos	branches: 1.12.6; Allow a single process to control multiple ttys (for pthreads using _REENTRANT) using multiple EditLine objects. Mostly from Preston A. Elder.
1.11	07-Aug-2003	agc	branches: 1.11.32; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.10	10-Mar-2003	christos	sig_t is non portable
1.9	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.8	09-Jan-2001	jdolecek	make constant arrays a const
1.7	04-Jan-2001	christos	fix dumb typo in signal setup [from OpenBSD] return -1 if alloc fails.
1.6	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.5	02-Jul-1999	simonb	More trailing white space.
1.4	06-Jul-1997	christos	Fix compiler warnings.
1.3	11-Apr-1997	christos	Portability fixes: __const -> const BADSIG -> SIG_ERR int flags -> u_int flags #if __STDC__ -> #ifdef __STDC__
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.11.32.1	24-Sep-2008	wrstuden	Merge in changes between wrstuden-revivesa-base-2 and wrstuden-revivesa-base-3.
1.12.6.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.27.2.1	02-Aug-2025	perseant	Sync with HEAD
1.11	09-May-2016	christos	s/protected/libedit_private/g
1.10	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.9	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.8	19-Feb-2009	christos	reset and redraw on sigcont. From Anon Ymous.
1.7	15-Feb-2009	christos	in order for read() to return EINTR we need to use sigaction, not signal, otherwise SA_RESTART is set.
1.6	12-Jul-2008	christos	branches: 1.6.6; Don't bother with SIGSTOP it cannot be caught or ignored. From Jess Thrysoee
1.5	07-Aug-2003	agc	branches: 1.5.32; Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.4	10-Mar-2003	christos	sig_t is non portable
1.3	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.5.32.1	18-Sep-2008	wrstuden	Sync with wrstuden-revivesa-base-2.
1.6.6.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.29	25-Apr-2023	christos	pass lint.
1.28	04-Feb-2023	christos	Remove unused stuff, and limit the scope of some of the used ones. (from des@freebsd)
1.27	09-May-2016	christos	s/protected/libedit_private/g
1.26	09-May-2016	christos	Instead of compiling all the source files together in one big file, use protected visibility to achieve the same effect.
1.25	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.24	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.23	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.22	16-Feb-2016	christos	- don't set _GNU_SOURCE. We are not supposed to make decisions for others. - don't special-case wcsdup() From Ingo Schwarze.
1.21	15-Feb-2016	christos	OpenBSD readline.c rev. 1.14 2015/02/06 23:21:58 millert use SIZE_MAX
1.20	15-Feb-2016	christos	forgot one fgetln define
1.19	15-Feb-2016	christos	change tests for fgetln.
1.18	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.17	28-Sep-2011	christos	include <wchar.h> if we don't have wcsdup()
1.16	29-Jul-2011	christos	kill ifdef notdef
1.15	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.14	28-Jul-2011	christos	more portability defines
1.13	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.12	31-Aug-2009	christos	delete defined(sun), it could be invaded in the user namespace. Suggested by mrg@
1.11	30-Aug-2009	christos	use __sun \|\| sun instead of _SunOS, from Jess Thrysoee
1.10	06-Feb-2009	sketch	SUNOS is spelt __SunOS. Add missing prototypes.
1.9	17-Jan-2004	christos	portability fixes.
1.8	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.7	19-Jun-2003	christos	provide a definition for __attribute__
1.6	10-Mar-2003	christos	- include cdefs.h early in the game - ifdef notdef the sunos stuff - no need for sig_t
1.5	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.4	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.3	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.2	02-Oct-1994	jtc	Added code so that POSIX.2 regular expresion functions are used if REGEX is defined, V8 regular expresion functions are used if REGEXP is defined, and BSD regular expression functions are used if neither are defined. And defined REGEX in sys.h so that programs using libedit don't have to link with libcompat.
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	branches: 1.1.1.1.2; libedit!
1.1.1.1.2.1	07-Oct-1994	mycroft	Update from trunk.
1.58	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.57	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.56	28-Dec-2009	christos	Reduce diff with tcsh's editor. No functional change intended.
1.55	31-Aug-2009	christos	delete defined(sun), it could be invaded in the user namespace. Suggested by mrg@
1.54	30-Aug-2009	christos	use __sun \|\| sun instead of _SunOS, from Jess Thrysoee
1.53	17-Jul-2009	christos	- off by one in the term.h case. - make code more similar to tcsh (if we want to handle wide chars, this is needed; for now it is a no-op)
1.52	31-Mar-2009	christos	cast to size_t to avoid sign / unsigned comparison warning.
1.51	31-Mar-2009	christos	Implement literal prompt sequences. Now someone can implement RL_PROMPT_START_LITERAL/RL_PROMPT_END_LITERAL :-)
1.50	15-Feb-2009	christos	pass lint on _LP64.
1.49	12-Feb-2009	sketch	More fixes for existing portability stuff.
1.48	06-Feb-2009	sketch	branches: 1.48.2; SUNOS is spelt __SunOS. Add missing prototypes.
1.47	10-Sep-2008	christos	Allow a single process to control multiple ttys (for pthreads using _REENTRANT) using multiple EditLine objects. Mostly from Preston A. Elder.
1.46	24-Nov-2006	christos	branches: 1.46.18; - Add more readline functions, enough for gdb-6.5 - Make el_get varyadic, and implement EL_GETTC. - XXX: the EL_SETTC api will change in the future.
1.45	18-Mar-2006	christos	Fix compilation.
1.44	18-Mar-2006	christos	Coverity CID 1668: Plug memory leak.
1.43	18-Mar-2006	christos	Coverity CID 806: Prevent NULL deref
1.42	06-Mar-2006	christos	Print the actual eofc, instead of ^D\b\b. Change internal character decoding to prevent buffer oveflows.
1.41	08-Aug-2005	christos	Spelling mistakes and comment errors (from FreeBSD via Stefan Farfeleder; many thanks)
1.40	22-May-2004	christos	branches: 1.40.2; fix memory leak; thanks to Logan Gabriel
1.39	17-Jan-2004	christos	portability fixes.
1.38	14-Sep-2003	christos	- provide enough hooks to compile gdb-5.3 - fix el_get(e, EL_TERMINAL, (char **))
1.37	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.36	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.35	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.34	08-Nov-2001	mycroft	In the `not what I asked for' department: Do NOT return an error from term_init() if term_set() fails. Otherwise el_init() barfs and libedit fails to work.
1.33	02-Nov-2001	christos	Finish initializing all the term data structures even if the terminal init fails. This makes editline work on dumb terminals again. Noted by mycroft. Oops, too agressive error checking.
1.32	23-Jan-2001	jdolecek	sprinkle couple const
1.31	10-Jan-2001	jdolecek	term_rebuffer_display(): set el->el_term.t_size.v to terminals height, not a magic value, so that e.g. el_display[] and el_vdisplay[] are not bigger than needed. Discussed with Christos Zoulas.
1.30	09-Jan-2001	jdolecek	term_beep(): use ordinary bell, don't attempt to use visual bell - if user wants visual instead of ordinary bell, they should set their environment appropriately
1.29	09-Jan-2001	jdolecek	make constant arrays a const
1.28	04-Jan-2001	christos	consistently check for allocation failures and return -1, if we could not get more memory.
1.27	30-Dec-2000	jdolecek	term_beep(): if terminal supports both visual and ordinary bell, 'ring' both
1.26	28-Nov-2000	jmc	Alloc the keys structure with A_K_NKEYS as the multiplier rather than a hardcoded value of 4. A_K_NKEYS is currently 6 and this mismatch was stomping memory when initializing the keys. (specifically gdb lost the exec file name if it was a long path name).
1.25	11-Nov-2000	christos	- add support for home and end keys. - improve debugging support
1.24	04-Oct-2000	sommerfeld	format string audit (silence warnings, save space)
1.23	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.22	02-Jun-2000	lukem	use strtol() (instead of atoi()) for sane error detection
1.21	19-Feb-2000	mycroft	branches: 1.21.2; Fix refresh glitches when using auto-margin.
1.20	20-Jan-2000	christos	Add support for automatic and magic margins (from tcsh) This makes the rightmost column usable on all programs that use editline.
1.19	24-Oct-1999	lukem	Fix pointer arithmatic (caused problems on LP64, including ftp dumping core when `edit' was turned off then on). Problem solved by David Huggins-Daines <dhd@eradicator.org>
1.18	04-Oct-1999	lukem	update post change to return value of tputs() third argument
1.17	02-Aug-1999	sommerfeld	branches: 1.17.2; Fix PR7685 (gdb under emacs prints spurious ^M and messes up terminal) plus a few bogons noted along the way: 1) Set EDIT_DISABLED if terminal type is emacs. 2) fix bug in NO_TTY mode which caused it to not notice CR or LF 3) implement EDIT_DISABLED within libedit to be somewhat like NO_TTY, except that a prompt is printed first.
1.16	02-Jul-1999	simonb	More trailing white space.
1.15	12-Dec-1998	christos	delint
1.14	20-May-1998	christos	remove term_beep hack.
1.13	30-Jan-1998	perry	update to lite-2 (just an sccsid change)
1.12	13-Nov-1997	thorpej	Un-"protect" term_beep() and rename it to __term_beep() to keep it out of the user's namespace. We need to do this because the readline emulation functions call term_beep(), but readline isn't built using the same (funky) namespace-protection method as the rest of libedit (it's included like a normal library object). Without this (fairly disgusting) hack, any program linked against libedit will fail to link with an unresolved reference to term_beep() if using an Elf toolchain (e.g. on the Alpha). XXX Why this doesn't happen with NetBSD's a.out toolchain is a mystery XXX to me, and I'm not sure I really want to know (given that a.out XXX _should_ break the same way as Elf does in this case).
1.11	13-Oct-1997	lukem	use <termcap.h> instead of "termcap.h" (which was repository copied to libterm)
1.10	06-Jul-1997	christos	Fix compiler warnings.
1.9	11-Apr-1997	christos	Return -1 if the terminal set operation resulted in dumb terminal settings.
1.8	23-Jan-1997	mrg	- convert unsafe strcpy(), strcat() and sprintf() to the `n' versions. - some KNF.
1.7	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.6	29-Apr-1995	christos	Fixed the key mapping code and reverted Charles' changes.
1.5	27-Apr-1995	mycroft	Remove dead code that can't possibly work.
1.4	07-Oct-1994	mycroft	Fix typo.
1.3	02-Oct-1994	mycroft	Don't even try to print out the name of the termcap file; it's hidden in libtermcap, and it normally uses the DB file anyway.
1.2	02-Oct-1994	mycroft	Nuke bogus baud rate conversion code.
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.2	30-Jan-1998	perry	import lite-2
1.1.1.1	06-May-1994	cgd	branches: 1.1.1.1.2; libedit!
1.1.1.1.2.1	07-Oct-1994	mycroft	Update from trunk.
1.17.2.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.21.2.1	23-Jun-2000	minoura	Sync w/ netbsd-1-5-base.
1.40.2.1	23-Apr-2009	snj	Apply patch (requested by msaitoh in ticket #2007): Coverity CID 1668: Plug memory leak when malloc() failed.
1.46.18.1	24-Sep-2008	wrstuden	Merge in changes between wrstuden-revivesa-base-2 and wrstuden-revivesa-base-3.
1.48.2.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.22	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.21	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.20	31-Mar-2009	christos	Implement literal prompt sequences. Now someone can implement RL_PROMPT_START_LITERAL/RL_PROMPT_END_LITERAL :-)
1.19	10-Sep-2008	christos	branches: 1.19.6; Allow a single process to control multiple ttys (for pthreads using _REENTRANT) using multiple EditLine objects. Mostly from Preston A. Elder.
1.18	24-Nov-2006	christos	branches: 1.18.18; - Add more readline functions, enough for gdb-6.5 - Make el_get varyadic, and implement EL_GETTC. - XXX: the EL_SETTC api will change in the future.
1.17	06-Mar-2006	christos	Print the actual eofc, instead of ^D\b\b. Change internal character decoding to prevent buffer oveflows.
1.16	15-Mar-2005	christos	Add missing define, needed for debugging (from Rob Rodgers)
1.15	14-Sep-2003	christos	- provide enough hooks to compile gdb-5.3 - fix el_get(e, EL_TERMINAL, (char **))
1.14	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.13	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.12	04-Jan-2001	christos	consistently check for allocation failures and return -1, if we could not get more memory.
1.11	11-Nov-2000	christos	- add support for home and end keys. - improve debugging support
1.10	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.9	20-Jan-2000	christos	Add support for automatic and magic margins (from tcsh) This makes the rightmost column usable on all programs that use editline.
1.8	04-Oct-1999	lukem	update post change to return value of tputs() third argument
1.7	02-Jul-1999	simonb	branches: 1.7.2; More trailing white space.
1.6	20-May-1998	christos	remove term_beep hack.
1.5	13-Nov-1997	thorpej	Un-"protect" term_beep() and rename it to __term_beep() to keep it out of the user's namespace. We need to do this because the readline emulation functions call term_beep(), but readline isn't built using the same (funky) namespace-protection method as the rest of libedit (it's included like a normal library object). Without this (fairly disgusting) hack, any program linked against libedit will fail to link with an unresolved reference to term_beep() if using an Elf toolchain (e.g. on the Alpha). XXX Why this doesn't happen with NetBSD's a.out toolchain is a mystery XXX to me, and I'm not sure I really want to know (given that a.out XXX _should_ break the same way as Elf does in this case).
1.4	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.3	29-Apr-1995	christos	Fixed the key mapping code and reverted Charles' changes.
1.2	27-Apr-1995	mycroft	Remove dead code that can't possibly work.
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.7.2.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.18.18.1	24-Sep-2008	wrstuden	Merge in changes between wrstuden-revivesa-base-2 and wrstuden-revivesa-base-3.
1.19.6.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.3	13-Oct-1997	mrg	termcap.h moves to libterm.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.46	04-Feb-2023	christos	Remove unused stuff, and limit the scope of some of the used ones. (from des@freebsd)
1.45	30-Oct-2022	christos	improvements in malloc/free handling.
1.44	09-Sep-2021	christos	Add casts to appease conversions between wchar_t and wint_t
1.43	10-Jul-2020	christos	Fix numeric variable handling in settc (lyzliyuzhi at 163 dot com)
1.42	31-May-2020	christos	use strlcpy() instead of strncpy() for gcc happiness
1.41	12-Nov-2019	christos	PR/54654: Soren Tempel: Make sure el_cursor.v < el_terminal.t_size.v when moving around.
1.40	15-Sep-2019	christos	Fix type and remove cast (Yuichiro NAITO/FreeBSD).
1.39	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.38	30-Jun-2019	christos	Add a comment explaining why we don't use DO here. Correct the attribution on the previous patch: The patch was from Jordan Lewis and the report from Raphael Poss.
1.37	29-Jun-2019	christos	PR/54329: Raphael Ross: According to https://www.gnu.org/software/termutils/\ manual/termcap-1.3/html_chapter/termcap_4.html#SEC23 the cursor move multiple escapes have undefined results when moving out of the screen. Stop using DO to move down multiple lines and use a loop of newlines instead.
1.36	12-Apr-2019	christos	PR/52359: Benjamin Lorenz: When resizing because of a signal save and restore the cursor position, since it does not change.
1.35	15-Feb-2019	christos	PR/53983: Jonathan Perkins: Fix types for readline compatibility
1.34	24-Nov-2018	christos	PR/53682: Jordan Lewis: use newlines instead of padded spaces when restoring multi-line histories.
1.33	27-Jun-2017	christos	branches: 1.33.4; 1.33.6; - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display.
1.32	09-May-2016	christos	branches: 1.32.8; s/protected/libedit_private/g
1.31	02-May-2016	christos	eliminate static buffer with custom resizing code.
1.30	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.29	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.28	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.27	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.26	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.25	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.24	22-Mar-2016	christos	put back NUL check (Ingo Schwarze)
1.23	02-Mar-2016	christos	PR/50880: David Binderman: Remove redundant code. While here, fix all debugging formats.
1.22	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.21	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.20	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.19	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.18	15-Feb-2016	christos	OpenBSD term.c rev. 1.7 2002/11/29 20:13:39 deraadt spelling
1.17	15-Feb-2016	christos	OpenBSD term.c rev. 1.13 2009/12/11 18:58:59 jacekm fix two memory leaks
1.16	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.15	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.14	30-May-2012	christos	don't include both term.h and termcap.h
1.13	24-Mar-2012	christos	From: Jilles Tjoelker: Add a mapping for the cursor delete key
1.12	18-Nov-2011	christos	Initialize termbuf (Kamil Dudka)
1.11	18-Nov-2011	christos	check for negative return of ct_visual_char (Kamil Dudka)
1.10	04-Oct-2011	christos	branches: 1.10.2; fixed warnings where wint_t is unsigned.
1.9	03-Oct-2011	christos	fix broken change (parenthesis in the wrong place). From Nirbhay Choubey
1.8	16-Aug-2011	christos	re-enable -Wconversion
1.7	29-Jul-2011	christos	pass -Wconversion
1.6	29-Jul-2011	christos	kill ifdef notdef
1.5	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.4	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.3	28-Jul-2011	christos	whitespace
1.2	28-Jul-2011	christos	Rename key to keymacro to avoid conflicts with term.h. The renaming of term to terminal was again to avoid conflicts with term.h. term.h is a moving namespace violation.
1.1	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.10.2.2	30-Oct-2012	yamt	sync with head
1.10.2.1	17-Apr-2012	yamt	sync with head
1.32.8.1	23-Jul-2017	snj	Pull up following revision(s) (requested by kre in ticket #102): lib/libedit/Makefile: 1.64-1.65 lib/libedit/editline.3: 1.94-1.96 lib/libedit/editrc.5: 1.33 lib/libedit/el.c: 1.93-1.94 lib/libedit/el.h: 1.42 lib/libedit/literal.c: 1.1-1.3 lib/libedit/literal.h: 1.1-1.2 lib/libedit/prompt.c: 1.27 lib/libedit/read.c: 1.103 lib/libedit/refresh.c: 1.52-1.54 lib/libedit/refresh.h: 1.11 lib/libedit/terminal.c: 1.33 Make the default editrc file be $EDITRC (from env) if set, falling back to $HOME/.editrc otherwise. Better support for this in sh coming. -- Include EDITRC in doc. -- mention the limitation of the literal sequence delimiter. -- - handle literal escape sequence printing. - factor out common code in allocation and freeing of the display. -- - add literal sequence handling. -- remove unused variable -- add literal escape sequence support, patterned after the tcsh ones. -- fix comment -- Fix an obvious, but almost invisible typo (avoid some core dumps). -- Allow wide characters (properly encoded as byte strings according to LC_CTYPE) to be (perhaps part of) the "invisible" characters in a prompt, or the required prompt character which follows the literal sequence (this character must be one with a printing column width >= 1). The literal indicator character (which is just a marker, and not printed anywhere) (the PSlit parameter in sh(1)) can also be a wide char (passed to libedit as a wchar_t, encoded as that by sh(1) or other applications that support this.) Note: this has currently only been tested with everything ascii (C locale). -- Remove workaround for ancient HTML generation code.
1.33.6.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.33.6.1	10-Jun-2019	christos	Sync with HEAD
1.33.4.1	26-Nov-2018	pgoyette	Sync with HEAD, resolve a couple of conflicts
1.9	09-May-2016	christos	s/protected/libedit_private/g
1.8	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.7	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.6	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.5	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.4	24-Mar-2012	christos	From: Jilles Tjoelker: Add a mapping for the cursor delete key
1.3	29-Jul-2011	christos	branches: 1.3.2; pass -Wconversion
1.2	28-Jul-2011	christos	Rename key to keymacro to avoid conflicts with term.h. The renaming of term to terminal was again to avoid conflicts with term.h. term.h is a moving namespace violation.
1.1	28-Jul-2011	christos	term -> terminal XXX: need to rename key_ too.
1.3.2.1	17-Apr-2012	yamt	sync with head
1.29	30-May-2023	christos	Fix some const qual (Piotr Pawel Stefaniak)
1.28	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.27	11-Apr-2016	christos	chartype cleanups from Ingo Schwarze: - The file tokenizer.c no longer uses chartype.h, so don't include the header. - The dummy definitions of ct_{de,en}code_string() for the NARROWCHAR case are only used in history.c, so move them there. - Now the whole content of chartype.h is for the wide character case only. So remove the NARROWCHAR ifdef and include the header only in the wide character case. - In chartype.h, move ct_encode_char() below the comment explaining it. - No more need for underscores before ct_{de,en}code_string(). - Make the conversion buffer resize functions private. They are only called from the decoding and encoding functions inside chartype.c, and no need can possibly arise to call them from anywhere else.
1.26	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.25	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.24	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.23	15-Feb-2016	christos	OpenBSD tokenizer.c rev. 1.8 2003/08/11 18:21:40 deraadt don't increase amax on realloc failure
1.22	30-Jan-2016	christos	Fix misplaced parentheses (Ingo Schwarze)
1.21	16-Aug-2011	christos	re-enable -Wconversion
1.20	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.19	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.18	03-Jan-2010	christos	rename historyw -> history_w for consistency. add wide tst code and make it the default.
1.17	30-Dec-2009	christos	Fix wide build, test it, but don't turn it on yet.
1.16	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.15	15-Feb-2009	christos	pass lint on _LP64.
1.14	05-Dec-2003	lukem	branches: 1.14.40; Tokenization function enhancements: * Make tok_init(), tok_end(), tok_reset(), tok_line() and tok_str() publically available in <histedit.h> * Documented the public functions in editline(3) * Renamed tok_line() -> tok_str() * Added new tok_line() which takes a "const LineInfo " instead of "const char " (the former has "cursor" information), and optionally return the argv index ("int cursorc") and offset within that index ("int cursorv"). This means that completion routines can use the tokenization code to crack the line and easily find which word the cursor is at. (mmm, context sensitive completion :) * Fixed TEST/test.c when using "continuation" lines (unmatched quote or \ at EOL), and added some more DEBUG messages including highlighting where the cursor is (with a `_').
1.13	18-Oct-2003	christos	change allocation policy in el_push to allocate the string itself. fix issues with strdup.
1.12	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.11	27-Oct-2002	christos	don't crash in memory shortage conditions.
1.10	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.9	31-Jan-2002	christos	don't forget to re-adjust the limit.
1.8	31-Jan-2002	christos	make pointer arithmetic more palatable.
1.7	04-Jan-2001	christos	consistently check for allocation failures and return -1, if we could not get more memory.
1.6	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.5	02-Jul-1999	simonb	More trailing white space.
1.4	12-Dec-1998	christos	delint
1.3	06-Jul-1997	christos	Fix compiler warnings.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.14.40.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.7	05-Dec-2003	lukem	Tokenization function enhancements: * Make tok_init(), tok_end(), tok_reset(), tok_line() and tok_str() publically available in <histedit.h> * Documented the public functions in editline(3) * Renamed tok_line() -> tok_str() * Added new tok_line() which takes a "const LineInfo " instead of "const char " (the former has "cursor" information), and optionally return the argv index ("int cursorc") and offset within that index ("int cursorv"). This means that completion routines can use the tokenization code to crack the line and easily find which word the cursor is at. (mmm, context sensitive completion :) * Fixed TEST/test.c when using "continuation" lines (unmatched quote or \ at EOL), and added some more DEBUG messages including highlighting where the cursor is (with a `_').
1.6	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.5	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.4	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.3	02-Jul-1999	simonb	More trailing white space.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.1	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.70	14-Jul-2021	christos	Via Jess Thrysoee, from Adrian Bunk: Fix libedit build on Linux/Alpha Alpha is the only Linux architecture that has SIGINFO: https://sources.debian.org/src/manpages/5.10-1/man7/signal.7/#L522 But even on Alpha Ctrl-T is not supported, and therefore no VSTATUS: https://sources.debian.org/src/manpages/5.10-1/man3/termios.3/#L603-L608 For consistency check both signal existence and character existence
1.69	31-May-2020	christos	use strlcpy() instead of strncpy() for gcc happiness
1.68	02-Dec-2018	christos	Add a couple more readline compat functions.
1.67	01-Jan-2018	christos	branches: 1.67.2; 1.67.4; Only FLUSH if we are ending libedit; DRAIN if we suspend for readline. This allows pasting multiline buffers (Gerry Swislow)
1.66	05-Sep-2017	christos	For readline emulation, don't reset the tty to "sane" (cooked) mode if we did not start this way. Also set and reset the tty on entry and exit from readline() since this is what readline does.
1.65	09-May-2016	christos	s/protected/libedit_private/g
1.64	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.63	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.62	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.61	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.60	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.59	22-Mar-2016	christos	Fix reversed condition in tty_end() (Ingo Schwarze) Also don't succeed if calling setup twice.
1.58	27-Feb-2016	christos	PR/50863: John Hein: libedit el_end() messes up term settings if piped Keep track if we initialized the tty, and only reset it if we did.
1.57	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.56	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.55	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.54	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.53	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.52	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.51	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.50	11-Feb-2016	christos	- Add some more Char casts - reduce ifdefs by providing empty defs for nls functions (Ingo Schwarze)
1.49	08-Dec-2015	gson	unbreak the build
1.48	08-Dec-2015	christos	If we did not setup the tty, don't reset it.
1.47	14-May-2015	christos	fix warnings on ubuntu 32 bit (Miki Rozloznik)
1.46	18-Jun-2014	christos	Add stdlib.h for abort() (Jess Thrysoee)
1.45	19-May-2014	christos	PR/48821: If called from tty_stty(), recalculate flags.
1.44	19-May-2014	christos	more tty modes refactoring, no functional change intended.
1.43	19-May-2014	christos	Factor out some common code (more to be done) from PR/48821
1.42	15-May-2012	christos	branches: 1.42.2; 1.42.8; save and restore the tty settings on entry and exit respectively. cleanup debugging printfs.
1.41	04-Oct-2011	christos	branches: 1.41.2; fixed warnings where wint_t is unsigned.
1.40	16-Aug-2011	christos	re-enable -Wconversion
1.39	29-Jul-2011	christos	kill ifdef notdef
1.38	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.37	28-Jul-2011	christos	Rename key to keymacro to avoid conflicts with term.h. The renaming of term to terminal was again to avoid conflicts with term.h. term.h is a moving namespace violation.
1.36	28-Jul-2011	christos	term -> terminal
1.35	28-Jan-2011	christos	fix pasto
1.34	27-Jan-2011	christos	don't turn on editing if stdout is not a tty.
1.33	18-Apr-2010	christos	branches: 1.33.2; ffs needs strings.h
1.32	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.31	22-Jul-2009	christos	Don't depend on side effects inside an assert From Michael Cook mcook at bbn dot com
1.30	16-Feb-2009	christos	fix sign compare issues.
1.29	15-Feb-2009	christos	pass lint on _LP64.
1.28	06-Feb-2009	sketch	branches: 1.28.2; Needs errno.h
1.27	10-Sep-2008	christos	Allow a single process to control multiple ttys (for pthreads using _REENTRANT) using multiple EditLine objects. Mostly from Preston A. Elder.
1.26	30-Jul-2008	christos	handle EINTR in the termios operations, reported by the GHC folks
1.25	18-Mar-2006	christos	branches: 1.25.20; Coverity CID 597: remove dead code.
1.24	18-Mar-2006	christos	Coverity CID 1216: Prevent negative index use.
1.23	01-Jun-2005	lukem	Don't use non-standard uint or u_int.
1.22	29-May-2005	christos	PR/25694: Luke Mewburn: Don't abuse unconstify'ing a string and writing to it, because you'll core dump. Also remove extra const that gives pain to the irix compiler.
1.21	13-Aug-2004	mycroft	Delete-previous-char and delete-next-char without an argument are not supposed to modify the yank buffer in Emacs. Make it so.
1.20	18-Oct-2003	christos	remove debugging printf.
1.19	18-Oct-2003	christos	Allow setty to set chars using char=value
1.18	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.17	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.16	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.15	17-May-2001	christos	PR/12963:Jason Waterman: Fix signed cast problems.
1.14	09-Jan-2001	jdolecek	make constant arrays a const
1.13	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.12	02-Aug-1999	sommerfeld	Fix PR7685 (gdb under emacs prints spurious ^M and messes up terminal) plus a few bogons noted along the way: 1) Set EDIT_DISABLED if terminal type is emacs. 2) fix bug in NO_TTY mode which caused it to not notice CR or LF 3) implement EDIT_DISABLED within libedit to be somewhat like NO_TTY, except that a prompt is printed first.
1.11	02-Jul-1999	simonb	More trailing white space.
1.10	07-Feb-1999	christos	PR/6957: Wolfgang Helbig: libedit swaps CR and LF control chars.
1.9	27-Sep-1998	christos	Obey incoming tty char settings.
1.8	20-May-1998	christos	change M_* constants to MD_* to avoid clashes with <stream.h>
1.7	30-Mar-1998	mrg	use int rather than char as an array index.
1.6	20-Oct-1997	scottr	Add support for DTR/CTS flow control, from Bill Studenmund.
1.5	09-Oct-1997	christos	PR/4211: Dave Huang: don't lose VSTATUS and VEOL and any other characters that are VDISABLED by default.
1.4	06-Jul-1997	christos	Fix compiler warnings.
1.3	11-Apr-1997	christos	Portability fixes: __const -> const BADSIG -> SIG_ERR int flags -> u_int flags #if __STDC__ -> #ifdef __STDC__
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.25.20.2	24-Sep-2008	wrstuden	Merge in changes between wrstuden-revivesa-base-2 and wrstuden-revivesa-base-3.
1.25.20.1	18-Sep-2008	wrstuden	Sync with wrstuden-revivesa-base-2.
1.28.2.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.33.2.1	08-Feb-2011	bouyer	Sync with HEAD
1.41.2.1	23-May-2012	yamt	sync with head.
1.42.8.1	10-Aug-2014	tls	Rebase.
1.42.2.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.67.4.1	10-Jun-2019	christos	Sync with HEAD
1.67.2.1	26-Dec-2018	pgoyette	Sync with HEAD, resolve a few conflicts
1.24	31-Jul-2021	andvar	s/dependend/dependent/
1.23	02-Dec-2018	christos	Add a couple more readline compat functions.
1.22	01-Jan-2018	christos	branches: 1.22.2; 1.22.4; Only FLUSH if we are ending libedit; DRAIN if we suspend for readline. This allows pasting multiline buffers (Gerry Swislow)
1.21	09-May-2016	christos	s/protected/libedit_private/g
1.20	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.19	27-Feb-2016	christos	PR/50863: John Hein: libedit el_end() messes up term settings if piped Keep track if we initialized the tty, and only reset it if we did.
1.18	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.17	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.16	16-Feb-2016	christos	No need to include "sys.h" from here; it is included from config.h
1.15	19-May-2014	christos	more tty modes refactoring, no functional change intended.
1.14	15-May-2012	christos	branches: 1.14.2; 1.14.8; save and restore the tty settings on entry and exit respectively. cleanup debugging printfs.
1.13	16-Aug-2011	christos	branches: 1.13.2; re-enable -Wconversion
1.12	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.11	01-Jun-2005	lukem	Don't use non-standard uint or u_int.
1.10	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.9	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.8	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.7	26-Sep-1999	lukem	apparantly need <unistd.h> for portable way of getting _POSIX_VDISABLE
1.6	02-Jul-1999	simonb	branches: 1.6.2; More trailing white space.
1.5	20-May-1998	christos	change M_* constants to MD_* to avoid clashes with <stream.h>
1.4	11-Apr-1997	christos	Don't allow CSWTCH to interfere with CSUSP on __SVR4 systems.
1.3	11-Apr-1997	christos	Portability fixes: __const -> const BADSIG -> SIG_ERR int flags -> u_int flags #if __STDC__ -> #ifdef __STDC__
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.6.2.1	27-Dec-1999	wrstuden	Pull up to last week's -current.
1.13.2.1	23-May-2012	yamt	sync with head.
1.14.8.1	10-Aug-2014	tls	Rebase.
1.14.2.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.22.4.1	10-Jun-2019	christos	Sync with HEAD
1.22.2.1	26-Dec-2018	pgoyette	Sync with HEAD, resolve a few conflicts
1.64	28-Aug-2021	christos	Respect $EDITOR when execution one (Baptiste Daroussin)
1.63	23-Jul-2019	christos	PR/54399: S�ren Tempel: Uninitialized memory access in libedit history. Initialize the buffer using calloc. While here change all malloc(a * sizeof(b)) to calloc(a, sizeof(b)). XXX: should fix realloc similarly.
1.62	09-May-2016	christos	branches: 1.62.16; s/protected/libedit_private/g
1.61	18-Apr-2016	christos	From Ingo Schwarze: * Replace fcns.c by a shorter and simpler func.h and include it only in the one file needing it, map.c. * Combine help.h and help.c into a simplified help.h and include it only in the one file needing it, map.c. * Check the very simple, static files editline.c, historyn.c, and tokenizern.c into CVS rather than needlessly generating them. * So we no longer autogenerate any C files. :-) * Shorten and simplify makelist by deleting the options -n, -e, -bc, and -m; the latter was unused and useless in the first place. * Move the declaration of el_func_t from fcns.h to the header actually needing it, map.h. Since that header is already included by el.h for unrelated reasons, that makes el_func_t just as globally available as before. * No longer include the simplified fcns.h into el.h, include it directly into the *.c files needing it.
1.60	11-Apr-2016	christos	Get rid of private/public; keep protected (Ingo Schwarze)
1.59	11-Apr-2016	christos	Char -> wchar_t from Ingo Schwarze.
1.58	11-Apr-2016	christos	more macro WIDECHAR undoing from Ingo Schwarze.
1.57	09-Apr-2016	christos	Change some 0's to NULL's from Pedro Giffuni
1.56	09-Apr-2016	christos	More WIDECHAR elimination (Ingo Schwarze)
1.55	02-Mar-2016	christos	PR/50880: David Binderman: Remove redundant code. While here, fix all debugging formats.
1.54	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.53	16-Feb-2016	christos	More header cleanups from Ingo Schwarze.
1.52	16-Feb-2016	christos	more include file cleanup (Ingo Schwarze)
1.51	16-Feb-2016	christos	From Ingo Scharze: Let "el.h" include everything needed for struct editline, and don't include that stuff multiple times. That also improves consistency, also avoids circular inclusions, and also makes it easier to follow what is going on, even though not quite as nice. But it seems like the best we can do...
1.50	16-Feb-2016	christos	cleanup chartype.h includes (Ingo Schwarze)
1.49	16-Feb-2016	christos	cleanup inclusion of histedit.h (Ingo Schwarze)
1.48	14-Feb-2016	christos	From Ingo Schwarze: As we have seen before, "histedit.h" can never get rid of including the <wchar.h> header because using the data types defined there is deeply ingrained in the public interfaces of libedit. Now POSIX unconditionally requires that <wchar.h> defines the type wint_t. Consequently, it can be used unconditionally, no matter whether WIDECHAR is active or not. Consequently, the #define Int is pointless. Note that removing it is not gratuitious churn. Auditing for integer signedness problems is already hard when only fundamental types like "int" and "unsigned" are involved. It gets very hard when types come into the picture that have platform-dependent signedness, like "char" and "wint_t". Adding yet another layer on top, changing both the signedness and the width in a platform- dependent way, makes auditing yet harder, which IMHO is really dangerous. Note that while removing the #define, i already found one bug caused by this excessive complication - in the function re_putc() in refresh.c. If WIDECHAR was defined, it printed an Int = wint_t value with %c. Fortunately, that bug only affects debugging, not production. The fix is contained in the patch. With WIDECHAR, this doesn't change anything. For the case without WIDECHAR, i checked that none of the places wants to store values that might not fit in wint_t. This only changes internal interfaces; public ones remain unchanged.
1.47	21-Oct-2015	christos	Use the full buffer for the conversion; ideally we should be dynamically allocating this. From Jilles Tjoelker
1.46	19-Oct-2015	christos	make sure we have space for NUL and NUL terminate buffer array (Jilles Tjoelker)
1.45	18-Jun-2014	christos	Don't depend on weak aliases to define the vi "alias" expansion function, provide an API instead to set it.
1.44	18-Jun-2014	christos	accomodate FreeBSD's flavor of weak references.
1.43	16-Jan-2012	christos	branches: 1.43.6; 1.43.16; PR/45843: Henning Petersen: Fix resource leak on error.
1.42	18-Nov-2011	christos	Off by one in allocation could cause buffer overflow (Kamil Dudka)
1.41	04-Oct-2011	christos	branches: 1.41.2; fixed warnings where wint_t is unsigned.
1.40	16-Aug-2011	christos	re-enable -Wconversion
1.39	29-Jul-2011	christos	pass -Wconversion
1.38	29-Jul-2011	christos	- fix unused params - unconditionalize vis.h
1.37	29-Jul-2011	christos	KNF return (\1); -> return \1;
1.36	28-Jul-2011	christos	kill ptr_t and ioctl_t, add * sizeof(*foo) to all allocations.
1.35	28-Jul-2011	christos	term -> terminal
1.34	22-Feb-2011	joerg	Introduce __weakref_visible to handle the different required visibility for weak references. GCC 4.2+ and Clang require static, older GCC wants extern. Change __weak_reference to include sym. This requires changes the existing users to not reuse the name of the symbol, but avoids further differences between GCC 4.1 and GCC 4.2+/clang.
1.33	17-Feb-2011	joerg	Deal with changes in the weak_reference semantic in GCC 4.2 and later.
1.32	23-Oct-2010	christos	branches: 1.32.2; fix fd leak found by Igor Zinovik
1.31	30-Dec-2009	christos	Wide character support (UTF-8) from Johny Mattsson; currently disabled.
1.30	21-Feb-2009	christos	more size_t stuff.
1.29	15-Feb-2009	christos	pass lint on _LP64.
1.28	06-Feb-2009	sketch	branches: 1.28.2; Portability fix.
1.27	22-Oct-2006	mrg	__weakref__ attribute can not be applied to anything when inside function scope, so, move the extern of get_alias_text outside vi_alias(). fixes build problems with GCC 4.1-20061021.
1.26	18-May-2006	christos	change __weak_extern to __weak_reference so that gcc4 works.
1.25	06-Mar-2006	christos	Print the actual eofc, instead of ^D\b\b. Change internal character decoding to prevent buffer oveflows.
1.24	10-Aug-2005	christos	Don't save the el->el_line.cursor over a cv_insert call and use it later because it might change. From Stefan Farfedeler.
1.23	09-Aug-2005	christos	Don't delete the current line in vi mode when typing 'yy'. From Stefan Farfeleder.
1.22	08-Aug-2005	christos	Spelling mistakes and comment errors (from FreeBSD via Stefan Farfeleder; many thanks)
1.21	25-Apr-2005	matt	Terminate the arglist with a NULL instead of 0. (Shuts up gcc4.x)
1.20	13-Aug-2004	mycroft	Delete-previous-char and delete-next-char without an argument are not supposed to modify the yank buffer in Emacs. Make it so.
1.19	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.18	19-Jun-2003	christos	From michael@moria.de: - use __attribute__((__unused__)) in arguments where appropriate. - some int -> size_t and char * to const char * conversions.
1.17	04-Jun-2003	matt	Only return CC_EOF if ^D (VEOF) was the only thing on the line. Otherwise beep and don't do anything else. This mimics the behavor of ^D outside in normal terminal mode. (^D in vi scrolls forwards and as such isn't appropriate to emulation)
1.16	10-Mar-2003	dsl	Put the __weak_extern() back inside vi_alias, but after the extern for get_alias_text().
1.15	10-Mar-2003	he	Move the __weak_extern() (ifdef'ed) outside of the vi_alias() function, so that this compiles again.
1.14	10-Mar-2003	christos	if no __weak_extern, don't even try to do vi_alias.
1.13	10-Mar-2003	christos	fix uninitialized variable. grr, I want gcc-3.3
1.12	15-Nov-2002	christos	PR/18995: David Laight: libedit fixes for posix conformant sh The posix 'sh' specification defines vi-mode editing quite tightly. The netbsd libedit code (used by sh to do this) was missing several features, there were also minor errors in others. Compare netbsd sh to the definition available from: http://www.opengroup.org/onlinepubs/007904975/utilities/sh.html In particular the following were not supported: U - undo all changes to line \| - goto column Y - yank to end of line y - yank # - comment out current line @ - take input from shell alias [1] G - goto numbered line in history buffer v - edit history line with vi _ - append word from last input line . - redo last command Other minor changes have also been made. [1] This needs the shell to define an appropriate routine to return the text of the alias. There is no requirement that such a function exist.
1.11	31-Oct-2002	christos	support for % command [matching parens/brackets/braces] on vi modes. From David Laight, thanks!
1.10	27-Oct-2002	christos	vi mode and memory fixes from david laight.
1.9	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.8	04-Sep-2000	lukem	convert to new style guide, which includes: - ansi prototypes & features (such as stdargs) - 8 space indents
1.7	02-Jul-1999	simonb	More trailing white space.
1.6	05-Feb-1999	christos	delint.
1.5	12-Dec-1998	christos	delint
1.4	03-Feb-1998	perry	remove obsolete register declarations
1.3	06-Jul-1997	christos	Fix compiler warnings.
1.2	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.28.2.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.32.2.1	05-Mar-2011	bouyer	Sync with HEAD
1.41.2.1	17-Apr-2012	yamt	sync with head
1.43.16.1	10-Aug-2014	tls	Rebase.
1.43.6.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.62.16.1	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.8	15-Oct-2017	abhinav	Add support for escaping special characters when doing filename completion. For instance if the file name is "foo bar": $ ls foo<TAB> should get autocompleted to: $ ls foo\ bar Works for similar other characters too, which need escaping. Also, add an accompanying test program to ensure the escaping is correct in various scenarios (within quotes, without quotes, with other special characeters) Thanks to Christos for reviews, help and feedback.
1.7	23-Mar-2016	christos	Start removing the WIDECHAR ifdefs; building without it has stopped working anyway. (Ingo Schwarze)
1.6	15-Feb-2016	christos	Compile with WIDECHAR the same way the main Makefile does (Ingo Schwarze)
1.5	03-Feb-2010	roy	Userland now builds and uses terminfo instead of termcap. OK: core@, jdc@
1.4	03-Jan-2010	christos	rename historyw -> history_w for consistency. add wide tst code and make it the default.
1.3	31-Aug-2006	rpaulo	Rename TEST/test.c to avoid problems when doing a cleandir on case-insensitive file systems. ok'ed christos.
1.2	05-Dec-2003	lukem	Tokenization function enhancements: * Make tok_init(), tok_end(), tok_reset(), tok_line() and tok_str() publically available in <histedit.h> * Documented the public functions in editline(3) * Renamed tok_line() -> tok_str() * Added new tok_line() which takes a "const LineInfo " instead of "const char " (the former has "cursor" information), and optionally return the argv index ("int cursorc") and offset within that index ("int cursorv"). This means that completion routines can use the tokenization code to crack the line and easily find which word the cursor is at. (mmm, context sensitive completion :) * Fixed TEST/test.c when using "continuation" lines (unmatched quote or \ at EOL), and added some more DEBUG messages including highlighting where the cursor is (with a `_').
1.1	16-Oct-2003	christos	I got tired of typing a cc line.
1.1	10-Sep-2021	christos	Add an LLVM fuzzing wrapper for the portable libedit (Christian Holler)
1.2	29-Feb-2016	christos	convert to 2 clause
1.1	16-Sep-2010	christos	unbreak readline history.
1.7	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.6	18-Jun-2014	christos	cast gotsig because it is long on some systems.
1.5	18-Apr-2010	christos	branches: 1.5.12; 1.5.22; \033 is more portable than \e still.
1.4	15-Apr-2010	christos	From Jess Thrysoee: call setlocale so we can test UTF-8
1.3	17-Jul-2009	christos	Use the proper prompt printing function.
1.2	31-Mar-2009	christos	Implement literal prompt sequences. Now someone can implement RL_PROMPT_START_LITERAL/RL_PROMPT_END_LITERAL :-)
1.1	31-Aug-2006	rpaulo	branches: 1.1.26; Rename TEST/test.c to avoid problems when doing a cleandir on case-insensitive file systems. ok'ed christos.
1.1.26.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.5.22.1	10-Aug-2014	tls	Rebase.
1.5.12.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.19	31-Aug-2006	rpaulo	Rename TEST/test.c to avoid problems when doing a cleandir on case-insensitive file systems. ok'ed christos.
1.18	01-Jun-2005	lukem	Don't use non-standard uint or u_int.
1.17	28-May-2005	lukem	fix for WARNS=3
1.16	18-May-2005	christos	Make signal handler safe. From Michael Knudsen, many thanks.
1.15	08-Dec-2003	lukem	update for function name change
1.14	05-Dec-2003	lukem	Tokenization function enhancements: * Make tok_init(), tok_end(), tok_reset(), tok_line() and tok_str() publically available in <histedit.h> * Documented the public functions in editline(3) * Renamed tok_line() -> tok_str() * Added new tok_line() which takes a "const LineInfo " instead of "const char " (the former has "cursor" information), and optionally return the argv index ("int cursorc") and offset within that index ("int cursorv"). This means that completion routines can use the tokenization code to crack the line and easily find which word the cursor is at. (mmm, context sensitive completion :) * Fixed TEST/test.c when using "continuation" lines (unmatched quote or \ at EOL), and added some more DEBUG messages including highlighting where the cursor is (with a `_').
1.13	07-Aug-2003	agc	Move UCB-licensed code from 4-clause to 3-clause licence. Patches provided by Joel Baker in PR 22280, verified by myself.
1.12	09-Aug-2002	soren	Remove extraneous \n's in {err,warn}{,x}.
1.11	23-Mar-2002	christos	make sure this compiles cleanly and works.
1.10	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.9	04-Sep-2000	lukem	convert to new style guide
1.8	21-Sep-1999	lukem	fix id header...
1.7	02-Jul-1999	simonb	More trailing white space.
1.6	20-May-1998	christos	Show the simpler way.
1.5	20-May-1998	christos	Fix for api changes.
1.4	23-Oct-1997	lukem	make this compile with the new libedit history()
1.3	11-Jan-1997	lukem	RCSid police editline first appeared in 4.4BSD not NetBSD1.0
1.2	15-Oct-1995	christos	Added history load and save to file functions.
1.1	06-May-1994	cgd	branches: 1.1.1; Initial revision
1.1.1.1	06-May-1994	cgd	libedit!
1.5	08-Sep-2019	abhinav	PR lib/54510: Fix file completion inside quotes which broke in rev 1.53 While there also fix handling character appending in the file completions when inside quotes. For example when inside a quote, if the completion is a directory then append a '/' but don't close the quote. On the other hand when inside a quote if the completion is a file name and it is the only match then we can close the quote.
1.4	31-Mar-2019	abhinav	Perform quoting of filename completions when there are multiple matches as well Quoting of special characters in filename completion was implemented for single match case, this enables it for multiple matches as well. For example: $ touch 'foo bar' $ touch 'foo baz' $ ls fo<TAB> autocompletes to => $ ls foo\ ba hitting <TAB> again shows: foo bar foo baz This required unescaping escape sequences generated during last completion in order to find the word to complete. While there, also update the test to include cases for multiple matches. Reviewed by christos
1.3	04-May-2018	abhinav	branches: 1.3.2; Handle filename autocompletion when the cursor is at a backslash or quote character For example, handle following case: $ touch 'foo bar' $ ls foo\<TAB> --> $ ls foo\ bar Also add test cases for this. Thanks to Christos for review
1.2	15-Oct-2017	abhinav	branches: 1.2.2; Add copyright and rcs header to the test program. Also, remove a test case which was not supposed to be there. (While that test case works with the changes I committed, testing that test case with the test program is not possible in its current form. I'm working on that.)
1.1	15-Oct-2017	abhinav	Add support for escaping special characters when doing filename completion. For instance if the file name is "foo bar": $ ls foo<TAB> should get autocompleted to: $ ls foo\ bar Works for similar other characters too, which need escaping. Also, add an accompanying test program to ensure the escaping is correct in various scenarios (within quotes, without quotes, with other special characeters) Thanks to Christos for reviews, help and feedback.
1.2.2.1	21-May-2018	pgoyette	Sync with HEAD
1.3.2.2	13-Apr-2020	martin	Mostly merge changes from HEAD upto 20200411
1.3.2.1	10-Jun-2019	christos	Sync with HEAD
1.8	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.7	18-Jun-2014	christos	cast gotsig because it is long on some systems.
1.6	18-Nov-2011	christos	branches: 1.6.6; 1.6.16; Initialize res (Kamil Dudka)
1.5	04-Aug-2011	christos	branches: 1.5.2; need err.h now
1.4	28-Jul-2011	christos	- check and don't cast malloc return - more wide function use
1.3	27-Jul-2011	christos	add history testing code.
1.2	18-Apr-2010	christos	\033 is more portable than \e still.
1.1	03-Jan-2010	christos	rename historyw -> history_w for consistency. add wide tst code and make it the default.
1.5.2.1	17-Apr-2012	yamt	sync with head
1.6.16.1	10-Aug-2014	tls	Rebase.
1.6.6.1	20-Aug-2014	tls	Rebase to HEAD as of a few days ago.
1.8	17-Feb-2016	christos	whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.7	03-Aug-2003	lukem	Add INCSYMLINKS to <bsd.inc.mk> and <bsd.kinc.mk>, and use that instead of SYMLINKS to install symlinked header files. INCSYMLINKS are installed with 'make includes'. This avoids using SYMLINKS and hacks with the 'linkinstall' target in <bsd.links.mk>, as linksinstall occurs in 'make install' and hacks to get it to occur in 'make includes' weren't robust, as seen in lib/libdes. Yet more improvements to bsd.README.
1.6	19-Aug-2002	lukem	More use of ${NETBSDSRCDIR}/some/path instead of ${.CURDIR}/../../some/path
1.5	12-Dec-2001	tv	Readability cleanups; MKfoo=no -> NOfoo.
1.4	16-May-2001	enami	Use relative path for symlink target so that it won't point outside of tree.
1.3	14-May-2001	jdolecek	Back off previous and do differently: Ensure that the <readline/history.h> link to <readline/readline.h> is installed too for 'make includes'. Also, avoid creating obj dir. This is finally real fix for lib/12929. XXX Is this right?
1.2	14-May-2001	jdolecek	Use <bsd.inc.mk>, instead of <bsd.prog.mk>, so that the readline.h header is installed during the 'make includes'. Fixes lib/12929 by Martin Husemann.
1.1	05-Jan-2001	jdolecek	Standard location of readline headers is /usr/include/readline/, so install them there. readline.h of libedit had to move to subdirectory 'readline', due to the way BSD makefiles work; this is better than potentially fragile Makefile hacks
1.55	25-Apr-2023	christos	pass lint.
1.54	21-Apr-2023	christos	Align types with readline-8.2 (wiz@)
1.53	19-Feb-2022	christos	PR/56695: Walter Lozano: Correct declaration of hook functions.
1.52	08-Feb-2022	christos	PR/56693: Walter Lozano: Add support for rl_delete_text and rl_set_key
1.51	31-Jan-2022	christos	PR/56622: Walter Lozano: Improve readline compatibility by adding rl_readline_state support.
1.50	14-Jan-2022	christos	Don't use __printflike, use the explicit attribute so that this works outside NetBSD.
1.49	14-Jan-2022	tnn	libedit: rl_message: annotate __printflike
1.48	11-Jan-2022	christos	PR/56618: Walter Lozano: Improve libedit compatibility with readline by implementing: rl_copy_text, rl_erase_empty_line, rl_message, rl_on_new_line, rl_replace_line, rl_restore_prompt, rl_save_prompt
1.47	21-Aug-2021	christos	Add more api to make gdb-11 happy (but not gdbtui as usual)
1.46	07-Jun-2019	christos	PR/54280: rl_completer_quote_characters should be const for readline compat
1.45	15-Feb-2019	christos	PR/53983: Jonathan Perkins: Fix types for readline compatibility
1.44	02-Dec-2018	christos	Add a couple more readline compat functions.
1.43	09-Jun-2018	christos	branches: 1.43.2; Provide more compatibility with readline headers; now python-3.6.5 works when changing 'readline' -> 'edit' in setup.py. Revert previous conditional setting of unbuffered.
1.42	01-Sep-2017	christos	branches: 1.42.2; PR/51517: Jay West: Tty settings not restored on exit PR/51518: Jay West: prompt is interleaved with client output Both these issues are caused by rl_restore_handler not DTRT; fix it so that it kills the internal libedit state completely. This is inefficient, but it works. Also fix: 1. add append_history()/H_NSAVE_FP 2. call the rl_startup_hook before printing the first prompt as documented. callint it from rl_initialize breaks python, because the callback ends up being invoked before the readline module is installed, and we end up dereferencing a NULL pointer. 3. add el_resize_terminal. With those changes, s/lreadline/ledit/g in python works.
1.41	28-Oct-2016	christos	export rl_done
1.40	24-Aug-2016	christos	more compatible with readline history functions.
1.39	17-Feb-2016	christos	branches: 1.39.2; whitespace and header sorting changes (Ingo Schwarze). No functional changes.
1.38	15-Feb-2016	christos	OpenBSD readline.c rev. 1.13 2015/01/13 08:33:12 reyk rl_set_keyboard_input_timeout() for readline 4.2 compat
1.37	02-Jun-2015	christos	remove duplicate declaration
1.36	02-Jun-2015	christos	Adjust API to a more modern readline (Ryo Onodera)
1.35	26-May-2015	christos	- fix types of rl_completion_entry_function and rl_add_defun - call update pos before completion to refresh the screen From Thomas Eriksson
1.34	28-May-2013	christos	expose rl_catch_signals and explain what we are doing.
1.33	15-May-2012	christos	branches: 1.33.2; Add rl_completion_word_break_hook from: http://gitweb.dragonflybsd.org/dragonfly.git/commitdiff/\ 5ec6a45fa762b8cbd0305ca06acb8780335a486a
1.32	16-Sep-2010	christos	branches: 1.32.6; unbreak readline history.
1.31	04-Aug-2010	christos	provide rl_on_newline
1.30	07-Sep-2009	christos	apply apple patches from: http://opensource.apple.com/source/libedit/libedit-11/patches/
1.29	31-Aug-2009	christos	delete defined(sun), it could be invaded in the user namespace. Suggested by mrg@
1.28	30-Aug-2009	christos	use __sun \|\| sun instead of _SunOS, from Jess Thrysoee
1.27	09-Jul-2009	christos	add stdio.h since we are using FILE. From Grant Erickson
1.26	31-Mar-2009	christos	implement RL_PROMPT_{START,END}_IGNORE
1.25	12-Feb-2009	sketch	More fixes for existing portability stuff.
1.24	05-Feb-2009	christos	branches: 1.24.2; add rl_set_prompt
1.23	29-Apr-2008	martin	Convert to new 2 clause license
1.22	04-Apr-2008	christos	branches: 1.22.2; Add rl_forced_update_display() from Gerry Swislow
1.21	12-Aug-2007	christos	patches from Axel Thimm
1.20	27-May-2007	christos	Add rl_completion_matches, fix remove_history
1.19	24-Nov-2006	christos	- Add more readline functions, enough for gdb-6.5 - Make el_get varyadic, and implement EL_GETTC. - XXX: the EL_SETTC api will change in the future.
1.18	21-Aug-2006	christos	Change to a 3 clause copyright after permission of the holders.
1.17	14-Jul-2005	christos	PR/30747: David N. Williams: libedit is missing remove_history() Added, please test.
1.16	11-Jun-2005	christos	PR/30500: Paul Shupak: Inconsistent definition of tilde_expand(). Provide a layer of indirection between the readline compatibility functions and our internal implementation, so that we have the freedom to change the function signature.
1.15	10-Jun-2005	christos	tilde expand should take a const argument.
1.14	27-May-2005	agc	Sync the alternative readline interface with reality: + the rl_callback_handler_install takes a pointer to a void function which has one char * argument (it's called that way in the readline emulation source, otherwise there's no way to pass the line buffer to the function which processes the line when EOL is encountered) + provide a prototype for that function signature and use it Makes the callback readline interface work now.
1.13	12-Apr-2005	christos	PR/29958: Peter Bex: add rl_variable_bind and rl_attempted_completion_over
1.12	08-Sep-2004	christos	branches: 1.12.2; make rl_inhibit_completion visible.
1.11	17-Jan-2004	christos	portability fixes.
1.10	27-Oct-2003	christos	Make readline csh-like history work. From Gerry Swislow <gerry at certif dot com>.
1.9	16-Oct-2003	christos	More libedit readline emulation functions from: Gerry Swislow <gerry at certif dot com>
1.8	26-Sep-2003	christos	Implement enough of readline's 4.0 async mode to make gdb happy. This is not complete yet, but it seems to work... This required to introduce an unbuffered mode to el_gets(), but that was a minor change.
1.7	15-Sep-2003	tron	Include "sys/ttydefaults.h" to get standard definition of "CTRL" macro which avoids clash with custom one.
1.6	15-Sep-2003	christos	kludge around CTRL redef.
1.5	14-Sep-2003	christos	Match the stupid function pointer declarations with actual readline's 4.0. This is gross.
1.4	14-Sep-2003	christos	Avoid pre-ansi warning.
1.3	14-Sep-2003	christos	- provide enough hooks to compile gdb-5.3 - fix el_get(e, EL_TERMINAL, (char **))
1.2	18-Mar-2002	christos	- constify; passes all gcc and lint strict checks. - add config.h [Jason Evans], to create a portable version of libedit that can be easily compiled on other OS's.
1.1	05-Jan-2001	jdolecek	Standard location of readline headers is /usr/include/readline/, so install them there. readline.h of libedit had to move to subdirectory 'readline', due to the way BSD makefiles work; this is better than potentially fragile Makefile hacks
1.12.2.1	28-May-2005	tron	Pull up revision 1.14 (requested by agc in ticket #353): Sync the alternative readline interface with reality: + the rl_callback_handler_install takes a pointer to a void function which has one char * argument (it's called that way in the readline emulation source, otherwise there's no way to pass the line buffer to the function which processes the line when EOL is encountered) + provide a prototype for that function signature and use it Makes the callback readline interface work now.
1.22.2.1	18-May-2008	yamt	sync with head.
1.24.2.1	13-May-2009	jym	Sync with HEAD. Third (and last) commit. See http://mail-index.netbsd.org/source-changes/2009/05/13/msg221222.html
1.32.6.2	22-May-2014	yamt	sync with head. for a reference, the tree before this commit was tagged as yamt-pagecache-tag8. this commit was splitted into small chunks to avoid a limitation of cvs. ("Protocol error: too many arguments")
1.32.6.1	23-May-2012	yamt	sync with head.
1.33.2.1	23-Jun-2013	tls	resync from head
1.39.2.1	04-Nov-2016	pgoyette	Sync with HEAD
1.42.2.2	26-Dec-2018	pgoyette	Sync with HEAD, resolve a few conflicts
1.42.2.1	25-Jun-2018	pgoyette	Sync with HEAD
1.43.2.1	10-Jun-2019	christos	Sync with HEAD

OpenGrok