tool changes

branches: 1.15.44; 1.15.46; 1.15.50;
Document non-literal formats

getdate is in libutil.

fix build with objdirs

Add -d option, similar to gnu date.

doc2html works here now, so remove NOHTML

doc2hmtl is broken for recursive Oo ... Oc at this time

Remove WARNS=1 from all the subdirectory Makefiles, and add it to Makefile.inc
now that all /bin has been cleaned.

Build with WARNS

convert to new RCS id conventions.

Merge 4.4-Lite version.

Minor tweak for cross-compiling.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

date(1) says:

STANDARDS
The date utility is expected to be compatible with IEEE Std 1003.2
(“POSIX.2”).

yet the format used for the date string arg is:

[[[[[[CC]yy]mm]dd]HH]MM[.SS]]

whereas POSIX demands

mmddHHMM[[CC]yy]

Why anyone would ever want to use that archaic form is incomprehensible
to me, yet, that is what is required.

Implement support for the POSIX format if POSIXLY_CORRECT is set in
the environment (in the full date(1) build only) in a rather crude
way that relies upon the user getting the format correct.

Tools builds are unaffected - setting the time is not supported there.
(It could be made to work, half setting the time, using -j, is possible
in tools builds, but pointless, and this cheap implementation uses
strptime() which is not necessarily available to a tools build.)

Mention this (very briefly) in the man page.

Add -U and -z options to date(1).

-U allows the (internal) TZ setting to be returned
to what it was when date started running.

-z zone sets TZ to be "zone" unless that is an empty
string, in which case it causes TZ to be removed from
the environment.

These can be used together to manipulate TZ around
parsing of a -d arg, to allow its parsing to be done
in a different zone than the one used for output, and
all unrelated to the initial TZ setting (or system default).

Note that these new options only appear in the first
synopsis form in both the man page and usage(), as -d
also appears only there, and these options make little
sense without also using -d.

This is a very simple change which amounts to no more
that a few setenv() and unsetenv() calls.

Note that the order in which the -u and -d options are given matters.

branches: 1.54.2;
date(1): sync two [[[[[CC]yy]mm]dd]HH]MM[.SS] instances

The markup was the same (modulo Li vs Cm for the dot before the
seconds), but use the same source markup grouping/layout in both to
make this fact more obvious.

date(1): minor markup fixes

Add -R option for displaying time in RFC 5322 format, similar to GNU date.

branches: 1.51.2;
Add -f option to set the time. From FreeBSD.

date(1): Use .Dl for one-liners. Same output is generated.

date(1): Fix the offset in previous.

date(1): add example for how to get seconds since the Epoch output

branches: 1.47.4; 1.47.6;
Remove trailing whitespace and Tn macro.

sync with tzcode-2018c

Remove workaround for ancient HTML generation code.

The range of the SS field is [0,60]

From the POSIX time.h man page:
"The formal definition of UTC does not permit double leap seconds, so all mention of double leap seconds has been removed, and the range shortened from the former [0,61] seconds seen in previous versions of POSIX."

Document the version date first appeared.
Bump date.

branches: 1.42.16;
- fix timed(8) references;
- fix capitalization;
- add reference to environ(7) in `SEE ALSO' section, as it is
mentioned in the text.

From patch by Bug Hunting.

branches: 1.41.6;
Remove trailing whitespace.

For the -d option, refer to parsedate(3) for examples.

Show that "seconds" is the expected argument for -r.
(I didn't bump the manual page date for this minor change. If I
should let me know.)

Drop trailing whitespace.

getdate -> parsedate

Add -d option, similar to gnu date.

Add -j flag to not actually change the clock, just parse the date given
(if any) and then display it in the format given. Matches similar -j
flag functionality in FreeBSD/OpenBSD.

Change requested by George Georgalis on netbsd-users.

While I'm here, disambiguate the letters in the date string, by using
the same option letters as used by strftime(3).

Add a '-a' option to set the time using adjtime(2) instead of settimeofday(2).
Fix usage so that the options which apply to dat/time setting are on the
correct line.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22249, verified by myself.

.Nm does not need a dummy argument ("") before punctuation or
for correct formatting of the SYNOPSIS any longer.

New policy: New sentences start on a new line.
Patches by Robert Elz <kre at munnari oz au>, with minimal changes by me.

Generate <>& symbolically. I'm avoiding .../dist/... directories for now.

Sort sections.

Whitespace nits.

xref settimeofday(2)

Document /etc/localtime, per PR 8099 by Charles Hannum.

Formatting cleanup: more fonts, less punctuation.

branches: 1.24.4;
make example cut&paste-ready

branches: 1.23.2;
Work around a spurious warning.

remove unnecessary comment

Make the documented parsing order for the date match reality.

Fix documentation of `-r'.

Allow a century to be specified.
Cut and paste some text from touch(1) about date parsing.

s/UCT/UTC/

branches: 1.17.2;
Fix .Nm usage.

fix broken timed(8) xrefs, noted by David Brownlee.

Add documentation of the fact that date does indeed understand years
after 2000 (years 00 through 68 are interpreted as meaning 2000-2068;
this should be updated sometime before 2068 :-)

nuke the BUGS section -- it was vax specific

remove obsolete -d and -t options -- kernel timezone handling is long gone

a a => a (close pr 2204).

ENVIRONMENTAL -> ENVIRONMENT

Sync with 4.4lite2

Use "utility" instead of "command". Modern definitions of these terms
are distinct (See POSIX.2 glossary).

A utility is a executable, script or shell builtin; while a command
can be any of those things plus lists, pipelines, compound commands
(if, for, while) and shell function definitions.

convert to new RCS id conventions.

Merge 4.4-Lite version.

Print correct default output format.
Note that date (with a suitable strftime()), is 1003.2 compliant.

1003.2: 4.15.4: The output always shall be terminated by a newline character.

Add RCS indentifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

date(1) says:

STANDARDS
The date utility is expected to be compatible with IEEE Std 1003.2
(“POSIX.2”).

yet the format used for the date string arg is:

[[[[[[CC]yy]mm]dd]HH]MM[.SS]]

whereas POSIX demands

mmddHHMM[[CC]yy]

Why anyone would ever want to use that archaic form is incomprehensible
to me, yet, that is what is required.

Implement support for the POSIX format if POSIXLY_CORRECT is set in
the environment (in the full date(1) build only) in a rather crude
way that relies upon the user getting the format correct.

Tools builds are unaffected - setting the time is not supported there.
(It could be made to work, half setting the time, using -j, is possible
in tools builds, but pointless, and this cheap implementation uses
strptime() which is not necessarily available to a tools build.)

Mention this (very briefly) in the man page.

Add -U and -z options to date(1).

-U allows the (internal) TZ setting to be returned
to what it was when date started running.

-z zone sets TZ to be "zone" unless that is an empty
string, in which case it causes TZ to be removed from
the environment.

These can be used together to manipulate TZ around
parsing of a -d arg, to allow its parsing to be done
in a different zone than the one used for output, and
all unrelated to the initial TZ setting (or system default).

Note that these new options only appear in the first
synopsis form in both the man page and usage(), as -d
also appears only there, and these options make little
sense without also using -d.

This is a very simple change which amounts to no more
that a few setenv() and unsetenv() calls.

PR lib/58674 (and I suspect, several others)

Stop expecting strptime(3) in the tools build, it is an XSI
function, and the tools builds do not set up the environment
to expect XSI functions to be available.

This means dropping support for the -f option in the tools date
(which shouldn't matter, -f sets the format for the string used
to set the time, which the tools date does not support), and
open coding parsing for the -d option in the tools build.

There should be no changes to the date(1) that is installed in /bin

As a (minor) benefit, the tools -d support has become somewhat
more flexible than the previous simple strptime() implementation
allowed, and also does better error checking (no more Feb 30 etc).

Note: no change to the usage message, if -f is passed to the tools
date it will elicit a usage which says that -f is supported...
(that's just laziness...)

isleap() is not a standard function/macro, so if none of the include
files have defined a macro of that name, define it ourselves.

branches: 1.66.2;
Handle -d %Y%m%d in the tools version. This is used in the release notes
Makefile.inc when BUILDID is specified. Consider providing parsedate(3)
in libcompat, but then we'd need yacc...

Add -R to usage

Add -R option for displaying time in RFC 5322 format, similar to GNU date.

branches: 1.63.2;
Add -f option to set the time. From FreeBSD.

tool changes

branches: 1.61.16; 1.61.18; 1.61.22;
Provide real error messages.
Inspired by PR 49169 from David H. Gutteridge, but a much broader patch :-)

Annotate dead functions

revert the revert and fix the code properly.

roll back to 20101210 -- the current version just hung if one tried
to set the date

branches: 1.57.2;
check return value of localtime.

- check return of strtoll
- misc cleanup

Simplify by using setenv() instead of putenv().

putenv() no longer accepts a const string

Don't blow up on date +''. Patch from Robert Elz in followup to PR 39392.

branches: 1.52.4;
Remove the \n and tabs from the __COPYRIGHT() strings.

branches: 1.51.4;
Use strtoll rather than strtol to read the -r argument, so it will work
for wider time_t. Also mark a timed-related y2038 lossage for whenever
someone figures out the best way to shoot timed.

branches: 1.50.4; 1.50.8;
Use size_t not int for strlen().

branches: 1.49.2;
getdate -> parsedate

getdate is in libutil.

Add -d option, similar to gnu date.

Add -j flag to not actually change the clock, just parse the date given
(if any) and then display it in the format given. Matches similar -j
flag functionality in FreeBSD/OpenBSD.

Change requested by George Georgalis on netbsd-users.

While I'm here, disambiguate the letters in the date string, by using
the same option letters as used by strftime(3).

PR/28450: David H. Gutteridge: date(1) does not validate its input and
accepts and processes impossible dates

Patch applied, thanks!

don't include program name when calling err()

PR/34055: Kevin Massey: Date uses a fixed 1K buffer for strftime
but does not check when strftime fails because the buffer is too small.
Our implementation of strftime does not NUL terminat the buffer
on failure (it does not have to; the results are undefined), so
garbage gets printed. Use a dynamically allocated buffer instead.

More minor cleanup:
- remove unnecessary prototype for main()
- use symbolic exit code
- KNF & wrap a long line

Minor code cleanup:

* remove unneeded cast
* use err() instead of perror() + exit()
* use symbolic exit code

Patch contributed by Liam J. Foy <liamfoy@sepulcrum.org> in private mail.

sprinkle a little const, and now everything compiles with WARNS=3

Set (and report) the new time (not the old one).
I broke it in rev 1.38, someone has finally noticed....

Add a '-a' option to set the time using adjtime(2) instead of settimeofday(2).
Fix usage so that the options which apply to dat/time setting are on the
correct line.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22249, verified by myself.

As discusses a little while back on tech-userlevel:

If stdout is a tty, use vis(3) to print any filenames to prevent garbage
from being printed if the filename contains control- or other non-printable
characters.

While here, sprinkle some EXIT_FAILURE and NOTREACHED where appropriate.

Use setprogname. Patch by Petri Koistinen.

ANSIfication and KNF fixes by Petri Koistinen in bin/13691.

setthetime() is prototyped static; make it so.

Use strtol() to convert the -r parameter so we can use alternate bases
for that input.

When setting the date, ask mktime() to work out whether we're in DST or not.
Seems to fix PR bin/8750.

branches: 1.30.4;
Don't declare 'extern opt*' getopt variables.

Make this slightly more portable by using putenv().

cast to unsigned char for is*()

GMT -> UTC

For the case where mktime() fails exit with "illegal time" error rather than
"illegal time format", which is misleading.

Be more retentive about use of NOTREACHED and noreturn.

Delint.

__AUDIT__ cleanup.

Need <time.h> for time(), strftime(), localtime() and mktime() prototypes.

Slight code rearrangement.

Don't validate the mmddhhmm.ss values here. mktime(3) does it.

Make the documented parsing order for the date match reality.

Allow a century to be specified.
Cut and paste some text from touch(1) about date parsing.

Use TM_YEAR_BASE to adjust tm_year, not 1900.

lint

branches: 1.15.2;
__RCSID()/__COPYRIGHT()/prototype police.

pull in util.h rather than prototypeing logtwmp() locally.

remove obsolete -d and -t options -- kernel timezone handling is long gone

kill use of register

Sync with 4.4lite2

clean up RCS Ids slightly, etc.

convert to new RCS id conventions.

Merge 4.4-Lite version.

Minor tweaks.

1003.2: 4.15.4: The output always shall be terminated by a newline character.

Add RCS identifiers.

date comamnd ignored seconds value. patch from <mhw@cs.brown.edu>

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

branches: 1.8.84;
getdate is in libutil.

no names in params.

Add -d option, similar to gnu date.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22249, verified by myself.

ANSIfication and KNF fixes by Petri Koistinen in bin/13691.

convert to new RCS id conventions.

Merge 4.4-Lite version.

branches: 1.1.1;
Initial revision

getdate is in libutil.

Add -d option, similar to gnu date.

branches: 1.30.46;
revert the revert and fix the code properly.

roll back to 20101210 -- the current version just hung if one tried
to set the date

branches: 1.28.2;
- check return of strtoll
- misc cleanup

Use strtoll rather than strtol to read the -r argument, so it will work
for wider time_t. Also mark a timed-related y2038 lossage for whenever
someone figures out the best way to shoot timed.

branches: 1.26.4; 1.26.8;
* Prevent out-of-bounds read tsptype if a packet with
an unknown msg_type is received.
* Check the received packet size.
* Use strncpy() instead of strlcpy() so that we don't
send gibberish from the stack.
* No need to bind().
* htons()/htonl() use uint16_t/uint32_t not u_short/u_long.

Adapt date(1), the netsettime function, to new return value from socket(2)
when the address family is unsupported.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22249, verified by myself.

strlcpy

select() -> poll()

fix WARNS=2

ANSIfication and KNF fixes by Petri Koistinen in bin/13691.

Use socklen_t; for compilers with strict type checking.

Fixed off-by-one error where HOSTNAME+1 characters was copied into
tsp_name that has size HOSTNAME.

#ifdef BSD4_4 for sin_len

Delint.

- use an array MAXHOSTNAMELEN+1 size to hold hostnames
- ensure hostname from gethostname() is nul-terminated in all cases
- minor KNF
- use MAXHOSTNAMELEN over various other values/defines
- be safe will buffers that hold hostnames

Allow a century to be specified.
Cut and paste some text from touch(1) about date parsing.

use IP_PORTRANGE_LOW sockopt rather than home-grown code

lint

branches: 1.11.2;
__RCSID()/__COPYRIGHT()/prototype police.

Sync with 4.4lite2

Fill in sin_len.

convert to new RCS id conventions.

Merge 4.4-Lite version.

typo fix

should bzero automatic sockaddr's before use

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision