funstack.awk revision 1.1.1.1
1 ### ==================================================================== 2 ### @Awk-file{ 3 ### author = "Nelson H. F. Beebe", 4 ### version = "1.00", 5 ### date = "09 October 1996", 6 ### time = "15:57:06 MDT", 7 ### filename = "journal-toc.awk", 8 ### address = "Center for Scientific Computing 9 ### Department of Mathematics 10 ### University of Utah 11 ### Salt Lake City, UT 84112 12 ### USA", 13 ### telephone = "+1 801 581 5254", 14 ### FAX = "+1 801 581 4148", 15 ### URL = "http://www.math.utah.edu/~beebe", 16 ### checksum = "25092 977 3357 26493", 17 ### email = "beebe (at] math.utah.edu (Internet)", 18 ### codetable = "ISO/ASCII", 19 ### keywords = "BibTeX, bibliography, HTML, journal table of 20 ### contents", 21 ### supported = "yes", 22 ### docstring = "Create a journal cover table of contents from 23 ### <at>Article{...} entries in a journal BibTeX 24 ### .bib file for checking the bibliography 25 ### database against the actual journal covers. 26 ### The output can be either plain text, or HTML. 27 ### 28 ### Usage: 29 ### bibclean -max-width 0 BibTeX-file(s) | \ 30 ### bibsort -byvolume | \ 31 ### awk -f journal-toc.awk \ 32 ### [-v HTML=nnn] [-v INDENT=nnn] \ 33 ### [-v BIBFILEURL=url] >foo.toc 34 ### 35 ### or if the bibliography is already sorted 36 ### by volume, 37 ### 38 ### bibclean -max-width 0 BibTeX-file(s) | \ 39 ### awk -f journal-toc.awk \ 40 ### [-v HTML=nnn] [-v INDENT=nnn] \ 41 ### [-v BIBFILEURL=url] >foo.toc 42 ### 43 ### A non-zero value of the command-line option, 44 ### HTML=nnn, results in HTML output instead of 45 ### the default plain ASCII text (corresponding 46 ### to HTML=0). The 47 ### 48 ### The INDENT=nnn command-line option specifies 49 ### the number of blanks to indent each logical 50 ### level of HTML. The default is INDENT=4. 51 ### INDENT=0 suppresses indentation. The INDENT 52 ### option has no effect when the default HTML=0 53 ### (plain text output) option is in effect. 54 ### 55 ### When HTML output is selected, the 56 ### BIBFILEURL=url command-line option provides a 57 ### way to request hypertext links from table of 58 ### contents page numbers to the complete BibTeX 59 ### entry for the article. These links are 60 ### created by appending a sharp (#) and the 61 ### citation label to the BIBFILEURL value, which 62 ### conforms with the practice of 63 ### bibtex-to-html.awk. 64 ### 65 ### The HTML output form may be useful as a more 66 ### compact representation of journal article 67 ### bibliography data than the original BibTeX 68 ### file provides. Of course, the 69 ### table-of-contents format provides less 70 ### information, and is considerably more 71 ### troublesome for a computer program to parse. 72 ### 73 ### When URL key values are provided, they will 74 ### be used to create hypertext links around 75 ### article titles. This supports journals that 76 ### provide article contents on the World-Wide 77 ### Web. 78 ### 79 ### For parsing simplicity, this program requires 80 ### that BibTeX 81 ### 82 ### key = "value" 83 ### 84 ### and 85 ### 86 ### @String{name = "value"} 87 ### 88 ### specifications be entirely contained on 89 ### single lines, which is readily provided by 90 ### the `bibclean -max-width 0' filter. It also 91 ### requires that bibliography entries begin and 92 ### end at the start of a line, and that 93 ### quotation marks, rather than balanced braces, 94 ### delimit string values. This is a 95 ### conventional format that again can be 96 ### guaranteed by bibclean. 97 ### 98 ### This program requires `new' awk, as described 99 ### in the book 100 ### 101 ### Alfred V. Aho, Brian W. Kernighan, and 102 ### Peter J. Weinberger, 103 ### ``The AWK Programming Language'', 104 ### Addison-Wesley (1988), ISBN 105 ### 0-201-07981-X, 106 ### 107 ### such as provided by programs named (GNU) 108 ### gawk, nawk, and recent AT&T awk. 109 ### 110 ### The checksum field above contains a CRC-16 111 ### checksum as the first value, followed by the 112 ### equivalent of the standard UNIX wc (word 113 ### count) utility output of lines, words, and 114 ### characters. This is produced by Robert 115 ### Solovay's checksum utility.", 116 ### } 117 ### ==================================================================== 118 119 BEGIN { initialize() } 120 121 /^ *@ *[Ss][Tt][Rr][Ii][Nn][Gg] *{/ { do_String(); next } 122 123 /^ *@ *[Pp][Rr][Ee][Aa][Mm][Bb][Ll][Ee]/ { next } 124 125 /^ *@ *[Aa][Rr][Tt][Ii][Cc][Ll][Ee]/ { do_Article(); next } 126 127 /^ *@/ { do_Other(); next } 128 129 /^ *author *= *\"/ { do_author(); next } 130 131 /^ *journal *= */ { do_journal(); next } 132 133 /^ *volume *= *\"/ { do_volume(); next } 134 135 /^ *number *= *\"/ { do_number(); next } 136 137 /^ *year *= *\"/ { do_year(); next } 138 139 /^ *month *= */ { do_month(); next } 140 141 /^ *title *= *\"/ { do_title(); next } 142 143 /^ *pages *= *\"/ { do_pages(); next } 144 145 /^ *URL *= *\"/ { do_URL(); next } 146 147 /^ *} *$/ { if (In_Article) do_end_entry(); next } 148 149 END { terminate() } 150 151 152 ######################################################################## 153 # NB: The programming conventions for variables in this program are: # 154 # UPPERCASE global constants and user options # 155 # Initialuppercase global variables # 156 # lowercase local variables # 157 # Any deviation is an error! # 158 ######################################################################## 159 160 161 function do_Article() 162 { 163 In_Article = 1 164 165 Citation_label = $0 166 sub(/^[^\{]*{/,"",Citation_label) 167 sub(/ *, *$/,"",Citation_label) 168 169 Author = "" 170 Title = "" 171 Journal = "" 172 Volume = "" 173 Number = "" 174 Month = "" 175 Year = "" 176 Pages = "" 177 Url = "" 178 } 179 180 181 function do_author() 182 { 183 Author = TeX_to_HTML(get_value($0)) 184 } 185 186 187 function do_end_entry( k,n,parts) 188 { 189 n = split(Author,parts," and ") 190 if (Last_number != Number) 191 do_new_issue() 192 for (k = 1; k < n; ++k) 193 print_toc_line(parts[k] " and", "", "") 194 Title_prefix = html_begin_title() 195 Title_suffix = html_end_title() 196 if (html_length(Title) <= (MAX_TITLE_CHARS + MIN_LEADERS)) # complete title fits on line 197 print_toc_line(parts[n], Title, html_begin_pages() Pages html_end_pages()) 198 else # need to split long title over multiple lines 199 do_long_title(parts[n], Title, html_begin_pages() Pages html_end_pages()) 200 } 201 202 203 function do_journal() 204 { 205 if ($0 ~ /[=] *"/) # have journal = "quoted journal name", 206 Journal = get_value($0) 207 else # have journal = journal-abbreviation, 208 { 209 Journal = get_abbrev($0) 210 if (Journal in String) # replace abbrev by its expansion 211 Journal = String[Journal] 212 } 213 gsub(/\\-/,"",Journal) # remove discretionary hyphens 214 } 215 216 217 function do_long_title(author,title,pages, last_title,n) 218 { 219 title = trim(title) # discard leading and trailing space 220 while (length(title) > 0) 221 { 222 n = html_breakpoint(title,MAX_TITLE_CHARS+MIN_LEADERS) 223 last_title = substr(title,1,n) 224 title = substr(title,n+1) 225 sub(/^ +/,"",title) # discard any leading space 226 print_toc_line(author, last_title, (length(title) == 0) ? pages : "") 227 author = "" 228 } 229 } 230 231 232 function do_month( k,n,parts) 233 { 234 Month = ($0 ~ /[=] *"/) ? get_value($0) : get_abbrev($0) 235 gsub(/[\"]/,"",Month) 236 gsub(/ *# *\\slash *# */," / ",Month) 237 gsub(/ *# *-+ *# */," / ",Month) 238 n = split(Month,parts," */ *") 239 Month = "" 240 for (k = 1; k <= n; ++k) 241 Month = Month ((k > 1) ? " / " : "") \ 242 ((parts[k] in Month_expansion) ? Month_expansion[parts[k]] : parts[k]) 243 } 244 245 246 function do_new_issue() 247 { 248 Last_number = Number 249 if (HTML) 250 { 251 if (Last_volume != Volume) 252 { 253 Last_volume = Volume 254 print_line(prefix(2) "<BR>") 255 } 256 html_end_toc() 257 html_begin_issue() 258 print_line(prefix(2) Journal "<BR>") 259 } 260 else 261 { 262 print_line("") 263 print_line(Journal) 264 } 265 266 print_line(strip_html(vol_no_month_year())) 267 268 if (HTML) 269 { 270 html_end_issue() 271 html_toc_entry() 272 html_begin_toc() 273 } 274 else 275 print_line("") 276 } 277 278 279 function do_number() 280 { 281 Number = get_value($0) 282 } 283 284 285 function do_Other() 286 { 287 In_Article = 0 288 } 289 290 291 function do_pages() 292 { 293 Pages = get_value($0) 294 sub(/--[?][?]/,"",Pages) 295 } 296 297 298 function do_String() 299 { 300 sub(/^[^\{]*\{/,"",$0) # discard up to and including open brace 301 sub(/\} *$/,"",$0) # discard from optional whitespace and trailing brace to end of line 302 String[get_key($0)] = get_value($0) 303 } 304 305 306 function do_title() 307 { 308 Title = TeX_to_HTML(get_value($0)) 309 } 310 311 312 function do_URL( parts) 313 { 314 Url = get_value($0) 315 split(Url,parts,"[,;]") # in case we have multiple URLs 316 Url = trim(parts[1]) 317 } 318 319 320 function do_volume() 321 { 322 Volume = get_value($0) 323 } 324 325 326 function do_year() 327 { 328 Year = get_value($0) 329 } 330 331 332 function get_abbrev(s) 333 { # return abbrev from ``key = abbrev,'' 334 sub(/^[^=]*= */,"",s) # discard text up to start of non-blank value 335 sub(/ *,? *$/,"",s) # discard trailing optional whitspace, quote, 336 # optional comma, and optional space 337 return (s) 338 } 339 340 341 function get_key(s) 342 { # return kay from ``key = "value",'' 343 sub(/^ */,"",s) # discard leading space 344 sub(/ *=.*$/,"",s) # discard everthing after key 345 346 return (s) 347 } 348 349 350 function get_value(s) 351 { # return value from ``key = "value",'' 352 sub(/^[^\"]*\" */,"",s) # discard text up to start of non-blank value 353 sub(/ *\",? *$/,"",s) # discard trailing optional whitspace, quote, 354 # optional comma, and optional space 355 return (s) 356 } 357 358 359 function html_accents(s) 360 { 361 if (index(s,"\\") > 0) # important optimization 362 { 363 # Convert common lower-case accented letters according to the 364 # table on p. 169 of in Peter Flynn's ``The World Wide Web 365 # Handbook'', International Thomson Computer Press, 1995, ISBN 366 # 1-85032-205-8. The official table of ISO Latin 1 SGML 367 # entities used in HTML can be found in the file 368 # /usr/local/lib/html-check/lib/ISOlat1.sgml (your path 369 # may differ). 370 371 gsub(/{\\\a}/, "\\à", s) 372 gsub(/{\\'a}/, "\\á", s) 373 gsub(/{\\[\^]a}/,"\\â", s) 374 gsub(/{\\~a}/, "\\ã", s) 375 gsub(/{\\\"a}/, "\\ä", s) 376 gsub(/{\\aa}/, "\\å", s) 377 gsub(/{\\ae}/, "\\æ", s) 378 379 gsub(/{\\c{c}}/,"\\ç", s) 380 381 gsub(/{\\\e}/, "\\è", s) 382 gsub(/{\\'e}/, "\\é", s) 383 gsub(/{\\[\^]e}/,"\\ê", s) 384 gsub(/{\\\"e}/, "\\ë", s) 385 386 gsub(/{\\\i}/, "\\ì", s) 387 gsub(/{\\'i}/, "\\í", s) 388 gsub(/{\\[\^]i}/,"\\î", s) 389 gsub(/{\\\"i}/, "\\ï", s) 390 391 # ignore eth and thorn 392 393 gsub(/{\\~n}/, "\\ñ", s) 394 395 gsub(/{\\\o}/, "\\ò", s) 396 gsub(/{\\'o}/, "\\ó", s) 397 gsub(/{\\[\^]o}/, "\\ô", s) 398 gsub(/{\\~o}/, "\\õ", s) 399 gsub(/{\\\"o}/, "\\ö", s) 400 gsub(/{\\o}/, "\\ø", s) 401 402 gsub(/{\\\u}/, "\\ù", s) 403 gsub(/{\\'u}/, "\\ú", s) 404 gsub(/{\\[\^]u}/,"\\û", s) 405 gsub(/{\\\"u}/, "\\ü", s) 406 407 gsub(/{\\'y}/, "\\ý", s) 408 gsub(/{\\\"y}/, "\\ÿ", s) 409 410 # Now do the same for upper-case accents 411 412 gsub(/{\\\A}/, "\\À", s) 413 gsub(/{\\'A}/, "\\Á", s) 414 gsub(/{\\[\^]A}/, "\\Â", s) 415 gsub(/{\\~A}/, "\\Ã", s) 416 gsub(/{\\\"A}/, "\\Ä", s) 417 gsub(/{\\AA}/, "\\Å", s) 418 gsub(/{\\AE}/, "\\Æ", s) 419 420 gsub(/{\\c{C}}/,"\\Ç", s) 421 422 gsub(/{\\\e}/, "\\È", s) 423 gsub(/{\\'E}/, "\\É", s) 424 gsub(/{\\[\^]E}/, "\\Ê", s) 425 gsub(/{\\\"E}/, "\\Ë", s) 426 427 gsub(/{\\\I}/, "\\Ì", s) 428 gsub(/{\\'I}/, "\\Í", s) 429 gsub(/{\\[\^]I}/, "\\Î", s) 430 gsub(/{\\\"I}/, "\\Ï", s) 431 432 # ignore eth and thorn 433 434 gsub(/{\\~N}/, "\\Ñ", s) 435 436 gsub(/{\\\O}/, "\\Ò", s) 437 gsub(/{\\'O}/, "\\Ó", s) 438 gsub(/{\\[\^]O}/, "\\Ô", s) 439 gsub(/{\\~O}/, "\\Õ", s) 440 gsub(/{\\\"O}/, "\\Ö", s) 441 gsub(/{\\O}/, "\\Ø", s) 442 443 gsub(/{\\\U}/, "\\Ù", s) 444 gsub(/{\\'U}/, "\\Ú", s) 445 gsub(/{\\[\^]U}/, "\\Û", s) 446 gsub(/{\\\"U}/, "\\Ü", s) 447 448 gsub(/{\\'Y}/, "\\Ý", s) 449 450 gsub(/{\\ss}/, "\\ß", s) 451 452 # Others not mentioned in Flynn's book 453 gsub(/{\\'\\i}/,"\\í", s) 454 gsub(/{\\'\\j}/,"j", s) 455 } 456 return (s) 457 } 458 459 460 function html_begin_issue() 461 { 462 print_line("") 463 print_line(prefix(2) "<HR>") 464 print_line("") 465 print_line(prefix(2) "<H1>") 466 print_line(prefix(3) "<A NAME=\"" html_label() "\">") 467 } 468 469 470 function html_begin_pages() 471 { 472 return ((HTML && (BIBFILEURL != "")) ? ("<A HREF=\"" BIBFILEURL "#" Citation_label "\">") : "") 473 } 474 475 476 function html_begin_pre() 477 { 478 In_PRE = 1 479 print_line("<PRE>") 480 } 481 482 483 function html_begin_title() 484 { 485 return ((HTML && (Url != "")) ? ("<A HREF=\"" Url "\">") : "") 486 } 487 488 489 function html_begin_toc() 490 { 491 html_end_toc() 492 html_begin_pre() 493 } 494 495 496 function html_body( k) 497 { 498 for (k = 1; k <= BodyLines; ++k) 499 print Body[k] 500 } 501 502 function html_breakpoint(title,maxlength, break_after,k) 503 { 504 # Return the largest character position in title AFTER which we 505 # can break the title across lines, without exceeding maxlength 506 # visible characters. 507 if (html_length(title) > maxlength) # then need to split title across lines 508 { 509 # In the presence of HTML markup, the initialization of 510 # k here is complicated, because we need to advance it 511 # until html_length(title) is at least maxlength, 512 # without invoking the expensive html_length() function 513 # too frequently. The need to split the title makes the 514 # alternative of delayed insertion of HTML markup much 515 # more complicated. 516 break_after = 0 517 for (k = min(maxlength,length(title)); k < length(title); ++k) 518 { 519 if (substr(title,k+1,1) == " ") 520 { # could break after position k 521 if (html_length(substr(title,1,k)) <= maxlength) 522 break_after = k 523 else # advanced too far, retreat back to last break_after 524 break 525 } 526 } 527 if (break_after == 0) # no breakpoint found by forward scan 528 { # so switch to backward scan 529 for (k = min(maxlength,length(title)) - 1; \ 530 (k > 0) && (substr(title,k+1,1) != " "); --k) 531 ; # find space at which to break title 532 if (k < 1) # no break point found 533 k = length(title) # so must print entire string 534 } 535 else 536 k = break_after 537 } 538 else # title fits on one line 539 k = length(title) 540 return (k) 541 } 542 543 544 545 function html_end_issue() 546 { 547 print_line(prefix(3) "</A>") 548 print_line(prefix(2) "</H1>") 549 } 550 551 552 function html_end_pages() 553 { 554 return ((HTML && (BIBFILEURL != "")) ? "</A>" : "") 555 } 556 557 558 function html_end_pre() 559 { 560 if (In_PRE) 561 { 562 print_line("</PRE>") 563 In_PRE = 0 564 } 565 } 566 567 568 function html_end_title() 569 { 570 return ((HTML && (Url != "")) ? "</A>" : "") 571 } 572 573 574 function html_end_toc() 575 { 576 html_end_pre() 577 } 578 579 580 function html_fonts(s, arg,control_word,k,level,n,open_brace) 581 { 582 open_brace = index(s,"{") 583 if (open_brace > 0) # important optimization 584 { 585 level = 1 586 for (k = open_brace + 1; (level != 0) && (k <= length(s)); ++k) 587 { 588 if (substr(s,k,1) == "{") 589 level++ 590 else if (substr(s,k,1) == "}") 591 level-- 592 } 593 594 # {...} is now found at open_brace ... (k-1) 595 for (control_word in Font_decl_map) # look for {\xxx ...} 596 { 597 if (substr(s,open_brace+1,length(control_word)+1) ~ \ 598 ("\\" control_word "[^A-Za-z]")) 599 { 600 n = open_brace + 1 + length(control_word) 601 arg = trim(substr(s,n,k - n)) 602 if (Font_decl_map[control_word] == "toupper") # arg -> ARG 603 arg = toupper(arg) 604 else if (Font_decl_map[control_word] != "") # arg -> <TAG>arg</TAG> 605 arg = "<" Font_decl_map[control_word] ">" arg "</" Font_decl_map[control_word] ">" 606 return (substr(s,1,open_brace-1) arg html_fonts(substr(s,k))) 607 } 608 } 609 for (control_word in Font_cmd_map) # look for \xxx{...} 610 { 611 if (substr(s,open_brace - length(control_word),length(control_word)) ~ \ 612 ("\\" control_word)) 613 { 614 n = open_brace + 1 615 arg = trim(substr(s,n,k - n)) 616 if (Font_cmd_map[control_word] == "toupper") # arg -> ARG 617 arg = toupper(arg) 618 else if (Font_cmd_map[control_word] != "") # arg -> <TAG>arg</TAG> 619 arg = "<" Font_cmd_map[control_word] ">" arg "</" Font_cmd_map[control_word] ">" 620 n = open_brace - length(control_word) - 1 621 return (substr(s,1,n) arg html_fonts(substr(s,k))) 622 } 623 } 624 } 625 return (s) 626 } 627 628 629 function html_header() 630 { 631 USER = ENVIRON["USER"] 632 if (USER == "") 633 USER = ENVIRON["LOGNAME"] 634 if (USER == "") 635 USER = "????" 636 "hostname" | getline HOSTNAME 637 "date" | getline DATE 638 ("ypcat passwd | grep '^" USER ":' | awk -F: '{print $5}'") | getline PERSONAL_NAME 639 if (PERSONAL_NAME == "") 640 ("grep '^" USER ":' /etc/passwd | awk -F: '{print $5}'") | getline PERSONAL_NAME 641 642 643 print "<!-- WARNING: Do NOT edit this file. It was converted from -->" 644 print "<!-- BibTeX format to HTML by journal-toc.awk version " VERSION_NUMBER " " VERSION_DATE " -->" 645 print "<!-- on " DATE " -->" 646 print "<!-- for " PERSONAL_NAME " (" USER "@" HOSTNAME ") -->" 647 print "" 648 print "" 649 print "<!DOCTYPE HTML public \"-//IETF//DTD HTML//EN\">" 650 print "" 651 print "<HTML>" 652 print prefix(1) "<HEAD>" 653 print prefix(2) "<TITLE>" 654 print prefix(3) Journal 655 print prefix(2) "</TITLE>" 656 print prefix(2) "<LINK REV=\"made\" HREF=\"mailto:" USER "@" HOSTNAME "\">" 657 print prefix(1) "</HEAD>" 658 print "" 659 print prefix(1) "<BODY>" 660 } 661 662 663 function html_label( label) 664 { 665 label = Volume "(" Number "):" Month ":" Year 666 gsub(/[^A-Za-z0-9():,;.\/\-]/,"",label) 667 return (label) 668 } 669 670 671 function html_length(s) 672 { # Return visible length of s, ignoring any HTML markup 673 if (HTML) 674 { 675 gsub(/<\/?[^>]*>/,"",s) # remove SGML tags 676 gsub(/&[A-Za-z0-9]+;/,"",s) # remove SGML entities 677 } 678 return (length(s)) 679 } 680 681 682 function html_toc() 683 { 684 print prefix(2) "<H1>" 685 print prefix(3) "Table of contents for issues of " Journal 686 print prefix(2) "</H1>" 687 print HTML_TOC 688 } 689 690 691 function html_toc_entry() 692 { 693 HTML_TOC = HTML_TOC " <A HREF=\"#" html_label() "\">" 694 HTML_TOC = HTML_TOC vol_no_month_year() 695 HTML_TOC = HTML_TOC "</A><BR>" "\n" 696 } 697 698 699 function html_trailer() 700 { 701 html_end_pre() 702 print prefix(1) "</BODY>" 703 print "</HTML>" 704 } 705 706 707 function initialize() 708 { 709 # NB: Update these when the program changes 710 VERSION_DATE = "[09-Oct-1996]" 711 VERSION_NUMBER = "1.00" 712 713 HTML = (HTML == "") ? 0 : (0 + HTML) 714 715 if (INDENT == "") 716 INDENT = 4 717 718 if (HTML == 0) 719 INDENT = 0 # indentation suppressed in ASCII mode 720 721 LEADERS = " . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ." 722 723 MAX_TITLE_CHARS = 36 # 36 produces a 79-char output line when there is 724 # just an initial page number. If this is 725 # increased, the LEADERS string may need to be 726 # lengthened. 727 728 MIN_LEADERS = 4 # Minimum number of characters from LEADERS 729 # required when leaders are used. The total 730 # number of characters that can appear in a 731 # title line is MAX_TITLE_CHARS + MIN_LEADERS. 732 # Leaders are omitted when the title length is 733 # between MAX_TITLE_CHARS and this sum. 734 735 MIN_LEADERS_SPACE = " " # must be at least MIN_LEADERS characters long 736 737 Month_expansion["jan"] = "January" 738 Month_expansion["feb"] = "February" 739 Month_expansion["mar"] = "March" 740 Month_expansion["apr"] = "April" 741 Month_expansion["may"] = "May" 742 Month_expansion["jun"] = "June" 743 Month_expansion["jul"] = "July" 744 Month_expansion["aug"] = "August" 745 Month_expansion["sep"] = "September" 746 Month_expansion["oct"] = "October" 747 Month_expansion["nov"] = "November" 748 Month_expansion["dec"] = "December" 749 750 Font_cmd_map["\\emph"] = "EM" 751 Font_cmd_map["\\textbf"] = "B" 752 Font_cmd_map["\\textit"] = "I" 753 Font_cmd_map["\\textmd"] = "" 754 Font_cmd_map["\\textrm"] = "" 755 Font_cmd_map["\\textsc"] = "toupper" 756 Font_cmd_map["\\textsl"] = "I" 757 Font_cmd_map["\\texttt"] = "t" 758 Font_cmd_map["\\textup"] = "" 759 760 Font_decl_map["\\bf"] = "B" 761 Font_decl_map["\\em"] = "EM" 762 Font_decl_map["\\it"] = "I" 763 Font_decl_map["\\rm"] = "" 764 Font_decl_map["\\sc"] = "toupper" 765 Font_decl_map["\\sf"] = "" 766 Font_decl_map["\\tt"] = "TT" 767 Font_decl_map["\\itshape"] = "I" 768 Font_decl_map["\\upshape"] = "" 769 Font_decl_map["\\slshape"] = "I" 770 Font_decl_map["\\scshape"] = "toupper" 771 Font_decl_map["\\mdseries"] = "" 772 Font_decl_map["\\bfseries"] = "B" 773 Font_decl_map["\\rmfamily"] = "" 774 Font_decl_map["\\sffamily"] = "" 775 Font_decl_map["\\ttfamily"] = "TT" 776 } 777 778 function min(a,b) 779 { 780 return (a < b) ? a : b 781 } 782 783 784 function prefix(level) 785 { 786 # Return a prefix of up to 60 blanks 787 788 if (In_PRE) 789 return ("") 790 else 791 return (substr(" ", \ 792 1, INDENT * level)) 793 } 794 795 796 function print_line(line) 797 { 798 if (HTML) # must buffer in memory so that we can accumulate TOC 799 Body[++BodyLines] = line 800 else 801 print line 802 } 803 804 805 function print_toc_line(author,title,pages, extra,leaders,n,t) 806 { 807 # When we have a multiline title, the hypertext link goes only 808 # on the first line. A multiline hypertext link looks awful 809 # because of long underlines under the leading indentation. 810 811 if (pages == "") # then no leaders needed in title lines other than last one 812 t = sprintf("%31s %s%s%s", author, Title_prefix, title, Title_suffix) 813 else # last title line, with page number 814 { 815 n = html_length(title) # potentially expensive 816 extra = n % 2 # extra space for aligned leader dots 817 if (n <= MAX_TITLE_CHARS) # then need leaders 818 leaders = substr(LEADERS, 1, MAX_TITLE_CHARS + MIN_LEADERS - extra - \ 819 min(MAX_TITLE_CHARS,n)) 820 else # title (almost) fills line, so no leaders 821 leaders = substr(MIN_LEADERS_SPACE,1, \ 822 (MAX_TITLE_CHARS + MIN_LEADERS - extra - n)) 823 t = sprintf("%31s %s%s%s%s%s %4s", \ 824 author, Title_prefix, title, Title_suffix, \ 825 (extra ? " " : ""), leaders, pages) 826 } 827 828 Title_prefix = "" # forget any hypertext 829 Title_suffix = "" # link material 830 831 # Efficency note: an earlier version accumulated the body in a 832 # single scalar like this: "Body = Body t". Profiling revealed 833 # this statement as the major hot spot, and the change to array 834 # storage made the program more than twice as fast. This 835 # suggests that awk might benefit from an optimization of 836 # "s = s t" that uses realloc() instead of malloc(). 837 if (HTML) 838 Body[++BodyLines] = t 839 else 840 print t 841 } 842 843 844 function protect_SGML_characters(s) 845 { 846 gsub(/&/,"\\&",s) # NB: this one MUST be first 847 gsub(/</,"\\<",s) 848 gsub(/>/,"\\>",s) 849 gsub(/\"/,"\\"",s) 850 return (s) 851 } 852 853 854 function strip_braces(s, k) 855 { # strip non-backslashed braces from s and return the result 856 857 return (strip_char(strip_char(s,"{"),"}")) 858 } 859 860 861 function strip_char(s,c, k) 862 { # strip non-backslashed instances of c from s, and return the result 863 k = index(s,c) 864 if (k > 0) # then found the character 865 { 866 if (substr(s,k-1,1) != "\\") # then not backslashed char 867 s = substr(s,1,k-1) strip_char(substr(s,k+1),c) # so remove it (recursively) 868 else # preserve backslashed char 869 s = substr(s,1,k) strip_char(s,k+1,c) 870 } 871 return (s) 872 } 873 874 875 function strip_html(s) 876 { 877 gsub(/<\/?[^>]*>/,"",s) 878 return (s) 879 } 880 881 882 function terminate() 883 { 884 if (HTML) 885 { 886 html_end_pre() 887 888 HTML = 0 # NB: stop line buffering 889 html_header() 890 html_toc() 891 html_body() 892 html_trailer() 893 } 894 } 895 896 897 function TeX_to_HTML(s, k,n,parts) 898 { 899 # First convert the four SGML reserved characters to SGML entities 900 if (HTML) 901 { 902 gsub(/>/, "\\>", s) 903 gsub(/</, "\\<", s) 904 gsub(/"/, "\\"", s) 905 } 906 907 gsub(/[$][$]/,"$$",s) # change display math to triple dollars for split 908 n = split(s,parts,/[$]/)# split into non-math (odd) and math (even) parts 909 910 s = "" 911 for (k = 1; k <= n; ++k) # unbrace non-math part, leaving math mode intact 912 s = s ((k > 1) ? "$" : "") \ 913 ((k % 2) ? strip_braces(TeX_to_HTML_nonmath(parts[k])) : \ 914 TeX_to_HTML_math(parts[k])) 915 916 gsub(/[$][$][$]/,"$$",s) # restore display math 917 918 return (s) 919 } 920 921 922 function TeX_to_HTML_math(s) 923 { 924 # Mostly a dummy for now, but HTML 3 could support some math translation 925 926 gsub(/\\&/,"\\&",s) # reduce TeX ampersands to SGML entities 927 928 return (s) 929 } 930 931 932 function TeX_to_HTML_nonmath(s) 933 { 934 if (index(s,"\\") > 0) # important optimization 935 { 936 gsub(/\\slash +/,"/",s) # replace TeX slashes with conventional ones 937 gsub(/ *\\emdash +/," --- ",s) # replace BibNet emdashes with conventional ones 938 gsub(/\\%/,"%",s) # reduce TeX percents to conventional ones 939 gsub(/\\[$]/,"$",s) # reduce TeX dollars to conventional ones 940 gsub(/\\#/,"#",s) # reduce TeX sharps to conventional ones 941 942 if (HTML) # translate TeX markup to HTML 943 { 944 gsub(/\\&/,"\\&",s) # reduce TeX ampersands to SGML entities 945 s = html_accents(s) 946 s = html_fonts(s) 947 } 948 else # plain ASCII text output: discard all TeX markup 949 { 950 gsub(/\\\&/, "\\&", s) # reduce TeX ampersands to conventional ones 951 952 gsub(/\\[a-z][a-z] +/,"",s) # remove TeX font changes 953 gsub(/\\[^A-Za-z]/,"",s) # remove remaining TeX control symbols 954 } 955 } 956 return (s) 957 } 958 959 960 function trim(s) 961 { 962 gsub(/^[ \t]+/,"",s) 963 gsub(/[ \t]+$/,"",s) 964 return (s) 965 } 966 967 968 function vol_no_month_year() 969 { 970 return ("Volume " wrap(Volume) ", Number " wrap(Number) ", " wrap(Month) ", " wrap(Year)) 971 } 972 973 974 function wrap(value) 975 { 976 return (HTML ? ("<STRONG>" value "</STRONG>") : value) 977 } 978
Indexes created Mon Mar 02 05:31:46 UTC 2026