https://perl5.git.perl.org / perl5.git / blame
| Commit | Line | Data |
|---|---|---|
| 4633a7c4 LW |
1 | #!/usr/local/bin/perl |
| 2 | ||
| 3 | use Config; | |
| 4 | use File::Basename qw(&basename &dirname); | |
| 5 | ||
| 6 | # List explicitly here the variables you want Configure to | |
| 7 | # generate. Metaconfig only looks for shell variables, so you | |
| 8 | # have to mention them as if they were shell variables, not | |
| 9 | # %Config entries. Thus you write | |
| 10 | # $startperl | |
| 11 | # to ensure Configure will look for $Config{startperl}. | |
| 12 | ||
| 13 | # This forces PL files to create target in same directory as PL file. | |
| 14 | # This is so that make depend always knows where to find PL derivatives. | |
| 15 | chdir(dirname($0)); | |
| 16 | ($file = basename($0)) =~ s/\.PL$//; | |
| 17 | $file =~ s/\.pl$// | |
| a53f8285 | 18 | if ($^O eq 'VMS' or $^O eq 'os2'); # "case-forgiving" |
| 4633a7c4 LW |
19 | |
| 20 | open OUT,">$file" or die "Can't create $file: $!"; | |
| 21 | ||
| 22 | print "Extracting $file (with variable substitutions)\n"; | |
| 23 | ||
| 24 | # In this section, perl variables will be expanded during extraction. | |
| 25 | # You can use $Config{...} to use Configure variables. | |
| 26 | ||
| 27 | print OUT <<"!GROK!THIS!"; | |
| 5f05dabc | 28 | $Config{startperl} |
| 29 | eval 'exec $Config{perlpath} -S \$0 \${1+"\$@"}' | |
| 30 | if \$running_under_some_shell; | |
| 5d94fbed AD |
31 | !GROK!THIS! |
| 32 | ||
| 4633a7c4 LW |
33 | # In the following, perl variables are not expanded during extraction. |
| 34 | ||
| 35 | print OUT <<'!NO!SUBS!'; | |
| 5d94fbed | 36 | # |
| 748a9306 LW |
37 | # pod2latex, version 1.1 |
| 38 | # by Taro Kawagish (kawagish@imslab.co.jp), Jan 11, 1995. | |
| 39 | # | |
| 40 | # pod2latex filters Perl pod documents to LaTeX documents. | |
| 41 | # | |
| 42 | # What pod2latex does: | |
| 43 | # 1. Pod file 'perl_doc_entry.pod' is filtered to 'perl_doc_entry.tex'. | |
| 44 | # 2. Indented paragraphs are translated into | |
| 45 | # '\begin{verbatim} ... \end{verbatim}'. | |
| 46 | # 3. '=head1 heading' command is translated into '\section{heading}' | |
| 47 | # 4. '=head2 heading' command is translated into '\subsection*{heading}' | |
| 48 | # 5. '=over N' command is translated into | |
| 49 | # '\begin{itemize}' if following =item starts with *, | |
| 50 | # '\begin{enumerate}' if following =item starts with 1., | |
| 51 | # '\begin{description}' if else. | |
| 52 | # (indentation level N is ignored.) | |
| 53 | # 6. '=item * heading' command is translated into '\item heading', | |
| 54 | # '=item 1. heading' command is translated into '\item heading', | |
| 55 | # '=item heading' command(other) is translated into '\item[heading]'. | |
| 56 | # 7. '=back' command is translated into | |
| 57 | # '\end{itemize}' if started with '\begin{itemize}', | |
| 58 | # '\end{enumerate}' if started with '\begin{enumerate}', | |
| 59 | # '\end{description}' if started with '\begin{description}'. | |
| 60 | # 8. other paragraphs are translated into strings with TeX special characters | |
| 61 | # escaped. | |
| 62 | # 9. In heading text, and other paragraphs, the following translation of pod | |
| 63 | # quotes are done, and then TeX special characters are escaped after that. | |
| 64 | # I<text> to {\em text\/}, | |
| 65 | # B<text> to {\bf text}, | |
| 66 | # S<text> to text1, | |
| 67 | # where text1 is a string with blank characters replaced with ~, | |
| 68 | # C<text> to {\tt text2}, | |
| 69 | # where text2 is a string with TeX special characters escaped to | |
| 70 | # obtain a literal printout, | |
| 71 | # E<text> (HTML escape) to TeX escaped string, | |
| 72 | # L<text> to referencing string as is done by pod2man, | |
| 73 | # F<file> to {\em file\/}, | |
| 74 | # Z<> to a null string, | |
| 75 | # 10. those headings are indexed: | |
| 76 | # '=head1 heading' => \section{heading}\index{heading} | |
| 77 | # '=head2 heading' => \subsection*{heading}\index{heading} | |
| 78 | # only when heading does not match frequent patterns such as | |
| 79 | # DESCRIPTION, DIAGNOSTICS,... | |
| 80 | # '=item heading' => \item{heading}\index{heading} | |
| 81 | # | |
| 82 | # Usage: | |
| 83 | # pod2latex perl_doc_entry.pod | |
| 84 | # this will write to a file 'perl_doc_entry.tex'. | |
| 85 | # | |
| 86 | # To LaTeX: | |
| 87 | # The following commands need to be defined in the preamble of the LaTeX | |
| 88 | # document: | |
| 89 | # \def\C++{{\rm C\kern-.05em\raise.3ex\hbox{\footnotesize ++}}} | |
| 90 | # \def\underscore{\leavevmode\kern.04em\vbox{\hrule width 0.4em height 0.3pt}} | |
| 91 | # and \parindent should be set zero: | |
| 92 | # \setlength{\parindent}{0pt} | |
| 93 | # | |
| 94 | # Note: | |
| 95 | # This script was written modifing pod2man. | |
| 96 | # | |
| 97 | # Bug: | |
| 98 | # If HTML escapes E<text> other than E<amp>,E<lt>,E<gt>,E<quot> are used | |
| 99 | # in C<>, translation will produce wrong character strings. | |
| 100 | # Translation of HTML escapes of various European accents might be wrong. | |
| 101 | ||
| 102 | ||
| 103 | $/ = ""; # record separator is blank lines | |
| 104 | # TeX special characters. | |
| 105 | ##$tt_ables = "!@*()-=+|;:'\"`,./?<>"; | |
| 106 | $backslash_escapables = "#\$%&{}_"; | |
| 107 | $backslash_escapables2 = "#\$%&{}"; # except _ | |
| 108 | ##$nonverbables = "^\\~"; | |
| 109 | ##$bracketesc = "[]"; | |
| 110 | ##@tex_verb_fences = unpack("aaaaaaaaa","|#@!*+?:;"); | |
| 111 | ||
| 112 | @head1_freq_patterns # =head1 patterns which need not be index'ed | |
| 113 | = ("AUTHOR","Author","BUGS","DATE","DESCRIPTION","DIAGNOSTICS", | |
| 114 | "ENVIRONMENT","EXAMPLES","FILES","INTRODUCTION","NAME","NOTE", | |
| 115 | "SEE ALSO","SYNOPSIS","WARNING"); | |
| 116 | ||
| 117 | $indent = 0; | |
| 118 | ||
| 119 | # parse the pods, produce LaTeX. | |
| 120 | ||
| 121 | open(POD,"<$ARGV[0]") || die "cant open $ARGV[0]"; | |
| 122 | ($pod=$ARGV[0]) =~ s/\.pod$//; | |
| 123 | open(LATEX,">$pod.tex"); | |
| 124 | &do_hdr(); | |
| 125 | ||
| 126 | $cutting = 1; | |
| 127 | while (<POD>) { | |
| 128 | if ($cutting) { | |
| 129 | next unless /^=/; | |
| 130 | $cutting = 0; | |
| 131 | } | |
| 132 | chop; | |
| 133 | length || (print LATEX "\n") && next; | |
| 134 | ||
| 135 | # translate indented lines as a verabatim paragraph | |
| 136 | if (/^\s/) { | |
| 137 | @lines = split(/\n/); | |
| 138 | print LATEX "\\begin{verbatim}\n"; | |
| 139 | for (@lines) { | |
| 140 | 1 while s | |
| 141 | {^( [^\t]* ) \t ( \t* ) } | |
| 142 | { $1 . ' ' x (8 - (length($1)%8) + 8*(length($2))) }ex; | |
| 143 | print LATEX $_,"\n"; | |
| 144 | } | |
| 145 | print LATEX "\\end{verbatim}\n"; | |
| 146 | next; | |
| 147 | } | |
| 148 | ||
| 149 | # preserve '=item' line with pod quotes as they are. | |
| 150 | if (/^=item/) { | |
| 151 | ($bareitem = $_) =~ s/^=item\s*//; | |
| 152 | } | |
| 153 | ||
| 154 | # check for things that'll hosed our noremap scheme; affects $_ | |
| 155 | &init_noremap(); | |
| 156 | ||
| 157 | # expand strings "func()" as pod quotes. | |
| 158 | if (!/^=item/) { | |
| 159 | # first hide pod escapes. | |
| 160 | # escaped strings are mapped into the ones with the MSB's on. | |
| 161 | s/([A-Z]<[^<>]*>)/noremap($1)/ge; | |
| 162 | ||
| 163 | # func() is a reference to a perl function | |
| 164 | s{\b([:\w]+\(\))}{I<$1>}g; | |
| 165 | # func(n) is a reference to a man page | |
| 166 | s{(\w+)(\([^\s,\051]+\))}{I<$1>$2}g; | |
| 167 | # convert simple variable references | |
| 168 | # s/([\$\@%][\w:]+)/C<$1>/g; | |
| 169 | # s/\$[\w:]+\[[0-9]+\]/C<$&>/g; | |
| 170 | ||
| 171 | if (m{ ([\-\w]+\([^\051]*?[\@\$,][^\051]*?\)) | |
| 172 | }x && $` !~ /([LCI]<[^<>]*|-)$/ && !/^=\w/) | |
| 173 | { | |
| 174 | warn "``$1'' should be a [LCI]<$1> ref"; | |
| 175 | } | |
| 176 | while (/(-[a-zA-Z])\b/g && $` !~ /[\w\-]$/) { | |
| 177 | warn "``$1'' should be [CB]<$1> ref"; | |
| 178 | } | |
| 179 | ||
| 180 | # put back pod quotes so we get the inside of <> processed; | |
| 181 | $_ = &clear_noremap($_); | |
| 182 | } | |
| 183 | ||
| 184 | ||
| 185 | # process TeX special characters | |
| 186 | ||
| 187 | # First hide HTML quotes E<> since they can be included in C<>. | |
| 188 | s/(E<[^<>]+>)/noremap($1)/ge; | |
| 189 | ||
| 190 | # Then hide C<> type literal quotes. | |
| 191 | # String inside of C<> will later be expanded into {\tt ..} strings | |
| 192 | # with TeX special characters escaped as needed. | |
| 193 | s/(C<[^<>]*>)/&noremap($1)/ge; | |
| 194 | ||
| 195 | # Next escape TeX special characters including other pod quotes B< >,... | |
| 196 | # | |
| 197 | # NOTE: s/re/&func($str)/e evaluates $str just once in perl5. | |
| 198 | # (in perl4 evaluation takes place twice before getting passed to func().) | |
| 199 | ||
| 200 | # - hyphen => --- | |
| 201 | s/(\S+)(\s+)-+(\s+)(\S+)/"$1".&noremap(" --- ")."$4"/ge; | |
| 202 | # '-', '--', "-" => '{\tt -}', '{\tt --}', "{\tt -}" | |
| 203 | ## s/("|')(\s*)(-+)(\s*)\1/&noremap("$1$2\{\\tt $3\}$4$1")/ge; | |
| 204 | ## changed Wed Jan 25 15:26:39 JST 1995 | |
| 205 | # '-', '--', "-" => '$-$', '$--$', "$-$" | |
| 206 | s/(\s+)(['"])(-+)([^'"\-]*)\2(\s+|[,.])/"$1$2".&noremap("\$$3\$")."$4$2$5"/ge; | |
| 207 | s/(\s+)(['"])([^'"\-]*)(-+)(\s*)\2(\s+|[,.])/"$1$2$3".&noremap("\$$4\$")."$5$2$6"/ge; | |
| 208 | # (--|-) => ($--$|$-$) | |
| 209 | s/(\s+)\((-+)([=@%\$\+\\\|\w]*)(-*)([=@%\$\+\\\|\w]*)\)(\s+|[,.])/"$1\(".&noremap("\$$2\$")."$3".&noremap("\$$4\$")."$5\)$6"/ge; | |
| 210 | # numeral - => $-$ | |
| 211 | s/(\(|[0-9]+|\s+)-(\s*\(?\s*[0-9]+)/&noremap("$1\$-\$$2")/ge; | |
| 212 | # -- in quotes => two separate - | |
| 213 | s/B<([^<>]*)--([^<>]*)>/&noremap("B<$1\{\\tt --\}$2>")/ge; | |
| 214 | ||
| 215 | # backslash escapable characters except _. | |
| 216 | s/([$backslash_escapables2])/&noremap("\\$1")/ge; | |
| 217 | s/_/&noremap("\\underscore{}")/ge; # a litle thicker than \_. | |
| 218 | # quote TeX special characters |, ^, ~, \. | |
| 219 | s/\|/&noremap("\$|\$")/ge; | |
| 220 | s/\^/&noremap("\$\\hat{\\hspace{0.4em}}\$")/ge; | |
| 221 | s/\~/&noremap("\$\\tilde{\\hspace{0.4em}}\$")/ge; | |
| 222 | s/\\/&noremap("\$\\backslash{}\$")/ge; | |
| 223 | # quote [ and ] to be used in \item[] | |
| 224 | s/([\[\]])/&noremap("{\\tt $1}")/ge; | |
| 225 | # characters need to be treated differently in TeX | |
| 226 | # keep * if an item heading | |
| 227 | s/^(=item[ \t]+)[*]((.|\n)*)/"$1" . &noremap("*") . "$2"/ge; | |
| 228 | s/[*]/&noremap("\$\\ast\$")/ge; # other * | |
| 229 | ||
| 230 | # hide other pod quotes. | |
| 231 | s/([ABD-Z]<[^<>]*>)/&noremap($1)/ge; | |
| 232 | ||
| 233 | # escape < and > as math strings, | |
| 234 | # now that we are done with hiding pod <> quotes. | |
| 235 | s/</&noremap("\$<\$")/ge; | |
| 236 | s/>/&noremap("\$>\$")/ge; | |
| 237 | ||
| 238 | # put it back so we get the <> processed again; | |
| 239 | $_ = &clear_noremap($_); | |
| 240 | ||
| 241 | ||
| 242 | # Expand pod quotes recursively: | |
| 243 | # (1) type face directives [BIFS]<[^<>]*> to appropriate TeX commands, | |
| 244 | # (2) L<[^<>]*> to reference strings, | |
| 245 | # (3) C<[^<>]*> to TeX literal quotes, | |
| 246 | # (4) HTML quotes E<> inside of C<> quotes. | |
| 247 | ||
| 248 | # Hide E<> again since they can be included in C<>. | |
| 249 | s/(E<[^<>]+>)/noremap($1)/ge; | |
| 250 | ||
| 251 | $maxnest = 10; | |
| 252 | while ($maxnest-- && /[A-Z]</) { | |
| 253 | ||
| 254 | # bold and italic quotes | |
| 255 | s/B<([^<>]*)>/"{\\bf $1}"/eg; | |
| 256 | s#I<([^<>]*)>#"{\\em $1\\/}"#eg; | |
| 257 | ||
| 258 | # files and filelike refs in italics | |
| 259 | s#F<([^<>]*)>#"{\\em $1\\/}"#eg; | |
| 260 | ||
| 261 | # no break quote -- usually we want C<> for this | |
| 262 | s/S<([^<>]*)>/&nobreak($1)/eg; | |
| 263 | ||
| 264 | # LREF: a manpage(3f) | |
| 265 | s:L<([a-zA-Z][^\s\/]+)(\([^\)]+\))?>:the {\\em $1\\/}$2 manpage:g; | |
| 266 | ||
| 267 | # LREF: an =item on another manpage | |
| 268 | s{ | |
| 269 | L<([^/]+)/([:\w]+(\(\))?)> | |
| 270 | } {the C<$2> entry in the I<$1> manpage}gx; | |
| 271 | ||
| 272 | # LREF: an =item on this manpage | |
| 273 | s{ | |
| 274 | ((?:L</([:\w]+(\(\))?)> | |
| 275 | (,?\s+(and\s+)?)?)+) | |
| 276 | } { &internal_lrefs($1) }gex; | |
| 277 | ||
| 278 | # LREF: a =head2 (head1?), maybe on a manpage, maybe right here | |
| 279 | # the "func" can disambiguate | |
| 280 | s{ | |
| 281 | L<(?:([a-zA-Z]\S+?) /)?"?(.*?)"?> | |
| 282 | }{ | |
| 283 | do { | |
| 284 | $1 # if no $1, assume it means on this page. | |
| 285 | ? "the section on I<$2> in the I<$1> manpage" | |
| 286 | : "the section on I<$2>" | |
| 287 | } | |
| 288 | }gex; | |
| 289 | ||
| 290 | s/Z<>/\\&/g; # the "don't format me" thing | |
| 291 | ||
| 292 | # comes last because not subject to reprocessing | |
| 293 | s{ | |
| 294 | C<([^<>]*)> | |
| 295 | }{ | |
| 296 | do { | |
| 297 | ($str = $1) =~ tr/\200-\377/\000-\177/; #normalize hidden stuff | |
| 298 | # expand HTML escapes if any; | |
| 299 | # WARNING: if HTML escapes other than E<amp>,E<lt>,E<gt>, | |
| 300 | # E<quot> are in C<>, they will not be printed correctly. | |
| 301 | $str = &expand_HTML_escapes($str); | |
| 302 | $strverb = &alltt($str); # Tex verbatim escape of a string. | |
| 303 | &noremap("$strverb"); | |
| 304 | } | |
| 305 | }gex; | |
| 306 | ||
| 307 | # if ( /C<([^<>]*)/ ) { | |
| 308 | # $str = $1; | |
| 309 | # if ($str !~ /\|/) { # if includes | | |
| 310 | # s/C<([^<>]*)>/&noremap("\\verb|$str|")/eg; | |
| 311 | # } else { | |
| 312 | # print STDERR "found \| in C<.*> at paragraph $.\n"; | |
| 313 | # # find a character not contained in $str to use it as a | |
| 314 | # # separator of the \verb | |
| 315 | # ($chars = $str) =~ s/(\W)/\\$1/g; | |
| 316 | # ## ($chars = $str) =~ s/([\$<>,\|"'\-^{}()*+?\\])/\\$1/g; | |
| 317 | # @fence = grep(!/[ $chars]/,@tex_verb_fences); | |
| 318 | # s/C<([^<>]*)>/&noremap("\\verb$fence[0]$str$fence[0]")/eg; | |
| 319 | # } | |
| 320 | # } | |
| 321 | } | |
| 322 | ||
| 323 | ||
| 324 | # process each pod command | |
| 325 | if (s/^=//) { # if a command | |
| 326 | s/\n/ /g; | |
| 327 | ($cmd, $rest) = split(' ', $_, 2); | |
| 328 | $rest =~ s/^\s*//; | |
| 329 | $rest =~ s/\s*$//; | |
| 330 | ||
| 331 | if (defined $rest) { | |
| 332 | &escapes; | |
| 333 | } | |
| 334 | ||
| 335 | $rest = &clear_noremap($rest); | |
| 336 | $rest = &expand_HTML_escapes($rest); | |
| 337 | ||
| 338 | if ($cmd eq 'cut') { | |
| 339 | $cutting = 1; | |
| 340 | $lastcmd = 'cut'; | |
| 341 | } | |
| 342 | elsif ($cmd eq 'head1') { # heading type 1 | |
| 343 | $rest =~ s/^\s*//; $rest =~ s/\s*$//; | |
| 344 | print LATEX "\n\\subsection*{$rest}"; | |
| 345 | # put index entry | |
| 346 | ($index = $rest) =~ s/^(An?\s+|The\s+)//i; # remove 'A' and 'The' | |
| 347 | # index only those heads not matching the frequent patterns. | |
| 348 | foreach $pat (@head1_freq_patterns) { | |
| 349 | if ($index =~ /^$pat/) { | |
| 350 | goto freqpatt; | |
| 351 | } | |
| 352 | } | |
| 353 | print LATEX "%\n\\index{$index}\n" if ($index); | |
| 354 | freqpatt: | |
| 355 | $lastcmd = 'head1'; | |
| 356 | } | |
| 357 | elsif ($cmd eq 'head2') { # heading type 2 | |
| 358 | $rest =~ s/^\s*//; $rest =~ s/\s*$//; | |
| 359 | print LATEX "\n\\subsubsection*{$rest}"; | |
| 360 | # put index entry | |
| 361 | ($index = $rest) =~ s/^(An?\s+|The\s+)//i; # remove 'A' and 'The' | |
| 362 | $index =~ s/^Example\s*[1-9][0-9]*\s*:\s*//; # remove 'Example :' | |
| 363 | print LATEX "%\n\\index{$index}\n" if ($index); | |
| 364 | $lastcmd = 'head2'; | |
| 365 | } | |
| 366 | elsif ($cmd eq 'over') { # 1 level within a listing environment | |
| 367 | push(@indent,$indent); | |
| 368 | $indent = $rest + 0; | |
| 369 | $lastcmd = 'over'; | |
| 370 | } | |
| 371 | elsif ($cmd eq 'back') { # 1 level out of a listing environment | |
| 372 | $indent = pop(@indent); | |
| 373 | warn "Unmatched =back\n" unless defined $indent; | |
| 374 | $listingcmd = pop(@listingcmd); | |
| 375 | print LATEX "\n\\end{$listingcmd}\n" if ($listingcmd); | |
| 376 | $lastcmd = 'back'; | |
| 377 | } | |
| 378 | elsif ($cmd eq 'item') { # an item paragraph starts | |
| 379 | if ($lastcmd eq 'over') { # if we have just entered listing env | |
| 380 | # see what type of list environment we are in. | |
| 381 | if ($rest =~ /^[0-9]\.?/) { # if numeral heading | |
| 382 | $listingcmd = 'enumerate'; | |
| 383 | } elsif ($rest =~ /^\*\s*/) { # if * heading | |
| 384 | $listingcmd = 'itemize'; | |
| 385 | } elsif ($rest =~ /^[^*]/) { # if other headings | |
| 386 | $listingcmd = 'description'; | |
| 387 | } else { | |
| 388 | warn "unknown list type for item $rest"; | |
| 389 | } | |
| 390 | print LATEX "\n\\begin{$listingcmd}\n"; | |
| 391 | push(@listingcmd,$listingcmd); | |
| 392 | } elsif ($lastcmd ne 'item') { | |
| 393 | warn "Illegal '=item' command without preceding 'over':"; | |
| 394 | warn "=item $bareitem"; | |
| 395 | } | |
| 396 | ||
| 397 | if ($listingcmd eq 'enumerate') { | |
| 398 | $rest =~ s/^[0-9]+\.?\s*//; # remove numeral heading | |
| 399 | print LATEX "\n\\item"; | |
| 400 | print LATEX "{\\bf $rest}" if $rest; | |
| 401 | } elsif ($listingcmd eq 'itemize') { | |
| 402 | $rest =~ s/^\*\s*//; # remove * heading | |
| 403 | print LATEX "\n\\item"; | |
| 404 | print LATEX "{\\bf $rest}" if $rest; | |
| 405 | } else { # description item | |
| 406 | print LATEX "\n\\item[$rest]"; | |
| 407 | } | |
| 408 | $lastcmd = 'item'; | |
| 409 | $rightafter_item = 'yes'; | |
| 410 | ||
| 411 | # check if the item heading is short or long. | |
| 412 | ($itemhead = $rest) =~ s/{\\bf (\S*)}/$1/g; | |
| 413 | if (length($itemhead) < 4) { | |
| 414 | $itemshort = "yes"; | |
| 415 | } else { | |
| 416 | $itemshort = "no"; | |
| 417 | } | |
| 418 | # write index entry | |
| 419 | if ($pod =~ "perldiag") { # skip 'perldiag.pod' | |
| 420 | goto noindex; | |
| 421 | } | |
| 422 | # strip out the item of pod quotes and get a plain text entry | |
| 423 | $bareitem =~ s/\n/ /g; # remove newlines | |
| 424 | $bareitem =~ s/\s*$//; # remove trailing space | |
| 425 | $bareitem =~ s/[A-Z]<([^<>]*)>/$1/g; # remove <> quotes | |
| 426 | ($index = $bareitem) =~ s/^\*\s+//; # remove leading '*' | |
| 427 | $index =~ s/^(An?\s+|The\s+)//i; # remove 'A' and 'The' | |
| 428 | $index =~ s/^\s*[1-9][0-9]*\s*[.]\s*$//; # remove numeral only | |
| 429 | $index =~ s/^\s*\w\s*$//; # remove 1 char only's | |
| 430 | # quote ", @ and ! with " to be used in makeindex. | |
| 431 | $index =~ s/"/""/g; # quote " | |
| 432 | $index =~ s/@/"@/g; # quote @ | |
| 433 | $index =~ s/!/"!/g; # quote ! | |
| 434 | ($rest2=$rest) =~ s/^\*\s+//; # remove * | |
| 435 | $rest2 =~ s/"/""/g; # quote " | |
| 436 | $rest2 =~ s/@/"@/g; # quote @ | |
| 437 | $rest2 =~ s/!/"!/g; # quote ! | |
| 438 | if ($pod =~ "(perlfunc|perlvar)") { # when doc is perlfunc,perlvar | |
| 439 | # take only the 1st word of item heading | |
| 440 | $index =~ s/^([^{}\s]*)({.*})?([^{}\s]*)\s+.*/\1\2\3/; | |
| 441 | $rest2 =~ s/^([^{}\s]*)({.*})?([^{}\s]*)\s+.*/\1\2\3/; | |
| 442 | } | |
| 443 | if ($index =~ /[A-Za-z\$@%]/) { | |
| 444 | # write \index{plain_text_entry@TeX_string_entry} | |
| 445 | print LATEX "%\n\\index{$index\@$rest2}%\n"; | |
| 446 | } | |
| 447 | noindex: | |
| 448 | ; | |
| 449 | } | |
| 450 | else { | |
| 451 | warn "Unrecognized directive: $cmd\n"; | |
| 452 | } | |
| 453 | } | |
| 454 | else { # if not command | |
| 455 | &escapes; | |
| 456 | $_ = &clear_noremap($_); | |
| 457 | $_ = &expand_HTML_escapes($_); | |
| 458 | ||
| 459 | # if the present paragraphs follows an =item declaration, | |
| 460 | # put a line break. | |
| 461 | if ($lastcmd eq 'item' && | |
| 462 | $rightafter_item eq 'yes' && $itemshort eq "no") { | |
| 463 | print LATEX "\\hfil\\\\"; | |
| 464 | $rightafter_item = 'no'; | |
| 465 | } | |
| 466 | print LATEX "\n",$_; | |
| 467 | } | |
| 468 | } | |
| 469 | ||
| 470 | print LATEX "\n"; | |
| 471 | close(POD); | |
| 472 | close(LATEX); | |
| 473 | ||
| 474 | ||
| 475 | ######################################################################### | |
| 476 | ||
| 477 | sub do_hdr { | |
| 478 | print LATEX "% LaTeX document produced by pod2latex from \"$pod.pod\".\n"; | |
| 479 | print LATEX "% The followings need be defined in the preamble of this document:\n"; | |
| 480 | print LATEX "%\\def\\C++{{\\rm C\\kern-.05em\\raise.3ex\\hbox{\\footnotesize ++}}}\n"; | |
| 481 | print LATEX "%\\def\\underscore{\\leavevmode\\kern.04em\\vbox{\\hrule width 0.4em height 0.3pt}}\n"; | |
| 482 | print LATEX "%\\setlength{\\parindent}{0pt}\n"; | |
| 483 | print LATEX "\n"; | |
| 484 | $podq = &escape_tex_specials("\U$pod\E"); | |
| 485 | print LATEX "\\section{$podq}%\n"; | |
| 486 | print LATEX "\\index{$podq}"; | |
| 487 | print LATEX "\n"; | |
| 488 | } | |
| 489 | ||
| 490 | sub nobreak { | |
| 491 | my $string = shift; | |
| 492 | $string =~ s/ +/~/g; # TeX no line break | |
| 493 | $string; | |
| 494 | } | |
| 495 | ||
| 496 | sub noremap { | |
| 497 | local($thing_to_hide) = shift; | |
| 498 | $thing_to_hide =~ tr/\000-\177/\200-\377/; | |
| 499 | return $thing_to_hide; | |
| 500 | } | |
| 501 | ||
| 502 | sub init_noremap { | |
| 18b0293d | 503 | # escape high bit characters in input stream |
| 504 | s/([\200-\377])/"E<".ord($1).">"/ge; | |
| 748a9306 LW |
505 | } |
| 506 | ||
| 507 | sub clear_noremap { | |
| 508 | local($tmp) = shift; | |
| 509 | $tmp =~ tr/\200-\377/\000-\177/; | |
| 510 | return $tmp; | |
| 511 | } | |
| 512 | ||
| 513 | sub expand_HTML_escapes { | |
| 514 | local($s) = $_[0]; | |
| 18b0293d | 515 | $s =~ s { E<((\d+)|([A-Za-z]+))> } |
| 748a9306 LW |
516 | { |
| 517 | do { | |
| 18b0293d | 518 | defined($2) |
| 519 | ? do { chr($2) } | |
| 520 | : | |
| 521 | exists $HTML_Escapes{$3} | |
| 522 | ? do { $HTML_Escapes{$3} } | |
| 748a9306 LW |
523 | : do { |
| 524 | warn "Unknown escape: $& in $_"; | |
| 525 | "E<$1>"; | |
| 526 | } | |
| 527 | } | |
| 528 | }egx; | |
| 529 | return $s; | |
| 530 | } | |
| 531 | ||
| 532 | sub escapes { | |
| 533 | # make C++ into \C++, which is to be defined as | |
| 534 | # \def\C++{{\rm C\kern-.05em\raise.3ex\hbox{\footnotesize ++}}} | |
| 535 | s/\bC\+\+/\\C++{}/g; | |
| 536 | } | |
| 537 | ||
| 538 | # Translate a string into a TeX \tt string to obtain a verbatim print out. | |
| 539 | # TeX special characters are escaped by \. | |
| 540 | # This can be used inside of LaTeX command arguments. | |
| 541 | # We don't use LaTeX \verb since it doesn't work inside of command arguments. | |
| 542 | sub alltt { | |
| 543 | local($str) = shift; | |
| 544 | # other chars than #,\,$,%,&,{,},_,\,^,~ ([ and ] included). | |
| 545 | $str =~ s/([^${backslash_escapables}\\\^\~]+)/&noremap("$&")/eg; | |
| 546 | # chars #,\,$,%,&,{,} => \# , ... | |
| 547 | $str =~ s/([$backslash_escapables2])/&noremap("\\$&")/eg; | |
| 548 | # chars _,\,^,~ => \char`\_ , ... | |
| 549 | $str =~ s/_/&noremap("\\char`\\_")/eg; | |
| 550 | $str =~ s/\\/&noremap("\\char`\\\\")/ge; | |
| 551 | $str =~ s/\^/\\char`\\^/g; | |
| 552 | $str =~ s/\~/\\char`\\~/g; | |
| 553 | ||
| 554 | $str =~ tr/\200-\377/\000-\177/; # put back | |
| 555 | $str = "{\\tt ".$str."}"; # make it a \tt string | |
| 556 | return $str; | |
| 557 | } | |
| 558 | ||
| 559 | sub escape_tex_specials { | |
| 560 | local($str) = shift; | |
| 561 | # other chars than #,\,$,%,&,{,}, _,\,^,~ ([ and ] included). | |
| 562 | # backslash escapable characters #,\,$,%,&,{,} except _. | |
| 563 | $str =~ s/([$backslash_escapables2])/&noremap("\\$1")/ge; | |
| 564 | $str =~ s/_/&noremap("\\underscore{}")/ge; # \_ is too thin. | |
| 565 | # quote TeX special characters |, ^, ~, \. | |
| 566 | $str =~ s/\|/&noremap("\$|\$")/ge; | |
| 567 | $str =~ s/\^/&noremap("\$\\hat{\\hspace{0.4em}}\$")/ge; | |
| 568 | $str =~ s/\~/&noremap("\$\\tilde{\\hspace{0.4em}}\$")/ge; | |
| 569 | $str =~ s/\\/&noremap("\$\\backslash{}\$")/ge; | |
| 570 | # characters need to be treated differently in TeX | |
| 571 | # * | |
| 572 | $str =~ s/[*]/&noremap("\$\\ast\$")/ge; | |
| 573 | # escape < and > as math string, | |
| 574 | $str =~ s/</&noremap("\$<\$")/ge; | |
| 575 | $str =~ s/>/&noremap("\$>\$")/ge; | |
| 576 | $str =~ tr/\200-\377/\000-\177/; # put back | |
| 577 | return $str; | |
| 578 | } | |
| 579 | ||
| 580 | sub internal_lrefs { | |
| 581 | local($_) = shift; | |
| 582 | ||
| 583 | s{L</([^>]+)>}{$1}g; | |
| 584 | my(@items) = split( /(?:,?\s+(?:and\s+)?)/ ); | |
| 585 | my $retstr = "the "; | |
| 586 | my $i; | |
| 587 | for ($i = 0; $i <= $#items; $i++) { | |
| 588 | $retstr .= "C<$items[$i]>"; | |
| 589 | $retstr .= ", " if @items > 2 && $i != $#items; | |
| 590 | $retstr .= " and " if $i+2 == @items; | |
| 591 | } | |
| 592 | $retstr .= " entr" . ( @items > 1 ? "ies" : "y" ) | |
| 593 | . " elsewhere in this document"; | |
| 594 | ||
| 595 | return $retstr; | |
| 596 | } | |
| 597 | ||
| 598 | # map of HTML escapes to TeX escapes. | |
| 599 | BEGIN { | |
| 600 | %HTML_Escapes = ( | |
| 601 | 'amp' => '&', # ampersand | |
| 602 | 'lt' => '<', # left chevron, less-than | |
| 603 | 'gt' => '>', # right chevron, greater-than | |
| 604 | 'quot' => '"', # double quote | |
| 605 | ||
| 606 | "Aacute" => "\\'{A}", # capital A, acute accent | |
| 607 | "aacute" => "\\'{a}", # small a, acute accent | |
| 608 | "Acirc" => "\\^{A}", # capital A, circumflex accent | |
| 609 | "acirc" => "\\^{a}", # small a, circumflex accent | |
| 610 | "AElig" => '\\AE', # capital AE diphthong (ligature) | |
| 611 | "aelig" => '\\ae', # small ae diphthong (ligature) | |
| 612 | "Agrave" => "\\`{A}", # capital A, grave accent | |
| 613 | "agrave" => "\\`{a}", # small a, grave accent | |
| 614 | "Aring" => '\\u{A}', # capital A, ring | |
| 615 | "aring" => '\\u{a}', # small a, ring | |
| 616 | "Atilde" => '\\~{A}', # capital A, tilde | |
| 617 | "atilde" => '\\~{a}', # small a, tilde | |
| 618 | "Auml" => '\\"{A}', # capital A, dieresis or umlaut mark | |
| 619 | "auml" => '\\"{a}', # small a, dieresis or umlaut mark | |
| 620 | "Ccedil" => '\\c{C}', # capital C, cedilla | |
| 621 | "ccedil" => '\\c{c}', # small c, cedilla | |
| 622 | "Eacute" => "\\'{E}", # capital E, acute accent | |
| 623 | "eacute" => "\\'{e}", # small e, acute accent | |
| 624 | "Ecirc" => "\\^{E}", # capital E, circumflex accent | |
| 625 | "ecirc" => "\\^{e}", # small e, circumflex accent | |
| 626 | "Egrave" => "\\`{E}", # capital E, grave accent | |
| 627 | "egrave" => "\\`{e}", # small e, grave accent | |
| 628 | "ETH" => '\\OE', # capital Eth, Icelandic | |
| 629 | "eth" => '\\oe', # small eth, Icelandic | |
| 630 | "Euml" => '\\"{E}', # capital E, dieresis or umlaut mark | |
| 631 | "euml" => '\\"{e}', # small e, dieresis or umlaut mark | |
| 632 | "Iacute" => "\\'{I}", # capital I, acute accent | |
| 633 | "iacute" => "\\'{i}", # small i, acute accent | |
| 634 | "Icirc" => "\\^{I}", # capital I, circumflex accent | |
| 635 | "icirc" => "\\^{i}", # small i, circumflex accent | |
| 636 | "Igrave" => "\\`{I}", # capital I, grave accent | |
| 637 | "igrave" => "\\`{i}", # small i, grave accent | |
| 638 | "Iuml" => '\\"{I}', # capital I, dieresis or umlaut mark | |
| 639 | "iuml" => '\\"{i}', # small i, dieresis or umlaut mark | |
| 640 | "Ntilde" => '\\~{N}', # capital N, tilde | |
| 641 | "ntilde" => '\\~{n}', # small n, tilde | |
| 642 | "Oacute" => "\\'{O}", # capital O, acute accent | |
| 643 | "oacute" => "\\'{o}", # small o, acute accent | |
| 644 | "Ocirc" => "\\^{O}", # capital O, circumflex accent | |
| 645 | "ocirc" => "\\^{o}", # small o, circumflex accent | |
| 646 | "Ograve" => "\\`{O}", # capital O, grave accent | |
| 647 | "ograve" => "\\`{o}", # small o, grave accent | |
| 648 | "Oslash" => "\\O", # capital O, slash | |
| 649 | "oslash" => "\\o", # small o, slash | |
| 650 | "Otilde" => "\\~{O}", # capital O, tilde | |
| 651 | "otilde" => "\\~{o}", # small o, tilde | |
| 652 | "Ouml" => '\\"{O}', # capital O, dieresis or umlaut mark | |
| 653 | "ouml" => '\\"{o}', # small o, dieresis or umlaut mark | |
| 654 | "szlig" => '\\ss', # small sharp s, German (sz ligature) | |
| 655 | "THORN" => '\\L', # capital THORN, Icelandic | |
| 656 | "thorn" => '\\l',, # small thorn, Icelandic | |
| 657 | "Uacute" => "\\'{U}", # capital U, acute accent | |
| 658 | "uacute" => "\\'{u}", # small u, acute accent | |
| 659 | "Ucirc" => "\\^{U}", # capital U, circumflex accent | |
| 660 | "ucirc" => "\\^{u}", # small u, circumflex accent | |
| 661 | "Ugrave" => "\\`{U}", # capital U, grave accent | |
| 662 | "ugrave" => "\\`{u}", # small u, grave accent | |
| 663 | "Uuml" => '\\"{U}', # capital U, dieresis or umlaut mark | |
| 664 | "uuml" => '\\"{u}', # small u, dieresis or umlaut mark | |
| 665 | "Yacute" => "\\'{Y}", # capital Y, acute accent | |
| 666 | "yacute" => "\\'{y}", # small y, acute accent | |
| 667 | "yuml" => '\\"{y}', # small y, dieresis or umlaut mark | |
| 668 | ); | |
| 669 | } | |
| 5d94fbed | 670 | !NO!SUBS! |
| 4633a7c4 LW |
671 | |
| 672 | close OUT or die "Can't close $file: $!"; | |
| 673 | chmod 0755, $file or die "Can't reset permissions for $file: $!\n"; | |
| 674 | exec("$Config{'eunicefix'} $file") if $Config{'eunicefix'} ne ':'; |