| 24 |
{ |
{ |
| 25 |
print TEMP "</PRE>\n" if ($inpre); |
print TEMP "</PRE>\n" if ($inpre); |
| 26 |
print TEMP "</P>\n"; |
print TEMP "</P>\n"; |
| 27 |
} |
} |
| 28 |
$inpara = $inpre = 0; |
$inpara = $inpre = 0; |
| 29 |
$wrotetext = 0; |
$wrotetext = 0; |
| 30 |
} |
} |
| 34 |
sub new_para { |
sub new_para { |
| 35 |
&end_para(); |
&end_para(); |
| 36 |
print TEMP "<P>\n"; |
print TEMP "<P>\n"; |
| 37 |
$inpara = 1; |
$inpara = 1; |
| 38 |
} |
} |
| 39 |
|
|
| 40 |
|
|
| 41 |
# Main program |
# Main program |
| 42 |
|
|
| 43 |
|
$innf = 0; |
| 44 |
$inpara = 0; |
$inpara = 0; |
| 45 |
$inpre = 0; |
$inpre = 0; |
| 46 |
$wrotetext = 0; |
$wrotetext = 0; |
| 49 |
|
|
| 50 |
while ($#ARGV >= 0 && $ARGV[0] =~ /^-/) |
while ($#ARGV >= 0 && $ARGV[0] =~ /^-/) |
| 51 |
{ |
{ |
| 52 |
$toc = 1 if $ARGV[0] eq "-toc"; |
$toc = 1 if $ARGV[0] eq "-toc"; |
| 53 |
shift; |
shift; |
| 54 |
} |
} |
| 55 |
|
|
| 56 |
# Initial output to STDOUT |
# Initial output to STDOUT |
| 57 |
|
|
| 65 |
<p> |
<p> |
| 66 |
Return to the <a href="index.html">PCRE index page</a>. |
Return to the <a href="index.html">PCRE index page</a>. |
| 67 |
</p> |
</p> |
| 68 |
<p> |
<p> |
| 69 |
This page is part of the PCRE HTML documentation. It was generated automatically |
This page is part of the PCRE HTML documentation. It was generated automatically |
| 70 |
from the original man page. If there is any nonsense in it, please consult the |
from the original man page. If there is any nonsense in it, please consult the |
| 71 |
man page, in case the conversion went wrong. |
man page, in case the conversion went wrong. |
| 72 |
<br> |
<br> |
| 73 |
End |
End |
| 74 |
|
|
| 75 |
print "<ul>\n" if ($toc); |
print "<ul>\n" if ($toc); |
| 78 |
|
|
| 79 |
while (<STDIN>) |
while (<STDIN>) |
| 80 |
{ |
{ |
| 81 |
# Handle lines beginning with a dot |
# Handle lines beginning with a dot |
| 82 |
|
|
| 83 |
if (/^\./) |
if (/^\./) |
| 84 |
{ |
{ |
| 85 |
if (/^\.br/) |
# Some of the PCRE man pages used to contain instances of .br. However, |
| 86 |
|
# they should have all been removed because they cause trouble in some |
| 87 |
|
# (other) automated systems that translate man pages to HTML. Complain if |
| 88 |
|
# we find .br or .in (another macro that is deprecated). |
| 89 |
|
|
| 90 |
|
if (/^\.br/ || /^\.in/) |
| 91 |
|
{ |
| 92 |
|
print STDERR "\n*** Deprecated macro encountered - rewrite needed\n"; |
| 93 |
|
print STDERR "*** $_\n"; |
| 94 |
|
die "*** Processing abandoned\n"; |
| 95 |
|
} |
| 96 |
|
|
| 97 |
|
# Instead of .br, relevent "literal" sections are enclosed in .nf/.fi. |
| 98 |
|
|
| 99 |
|
elsif (/^\.nf/) |
| 100 |
|
{ |
| 101 |
|
$innf = 1; |
| 102 |
|
} |
| 103 |
|
|
| 104 |
|
elsif (/^\.fi/) |
| 105 |
{ |
{ |
| 106 |
print TEMP "<br>\n" if $wrotetext; |
$innf = 0; |
| 107 |
} |
} |
| 108 |
|
|
| 109 |
# Handling .sp is subtle. If it is inside a literal section, do nothing if |
# Handling .sp is subtle. If it is inside a literal section, do nothing if |
| 110 |
# the next line is a non literal text line; similarly, if not inside a |
# the next line is a non literal text line; similarly, if not inside a |
| 111 |
# literal section, do nothing if a literal follows. The point being that |
# literal section, do nothing if a literal follows. The point being that |
| 112 |
# the <pre> and </pre> that delimit literal sections will do the spacing. |
# the <pre> and </pre> that delimit literal sections will do the spacing. |
| 113 |
# Always skip if no previous output. |
# Always skip if no previous output. |
| 114 |
|
|
| 115 |
elsif (/^\.sp/) |
elsif (/^\.sp/) |
| 116 |
{ |
{ |
| 117 |
if ($wrotetext) |
if ($wrotetext) |
| 119 |
$_ = <STDIN>; |
$_ = <STDIN>; |
| 120 |
if ($inpre) |
if ($inpre) |
| 121 |
{ |
{ |
| 122 |
print TEMP "\n" if (/^[\s.]/); |
print TEMP "\n" if (/^[\s.]/); |
| 123 |
} |
} |
| 124 |
else |
else |
| 125 |
{ |
{ |
| 126 |
print TEMP "<br>\n<br>\n" if (!/^[\s.]/); |
print TEMP "<br>\n<br>\n" if (!/^[\s.]/); |
| 127 |
} |
} |
| 128 |
redo; # Now process the lookahead line we just read |
redo; # Now process the lookahead line we just read |
| 129 |
} |
} |
| 130 |
} |
} |
| 131 |
elsif (/^\.TP/ || /^\.PP/ || /^\.P/) |
elsif (/^\.TP/ || /^\.PP/ || /^\.P/) |
| 132 |
{ |
{ |
| 133 |
&new_para(); |
&new_para(); |
| 134 |
} |
} |
| 135 |
elsif (/^\.SH\s*("?)(.*)\1/) |
elsif (/^\.SH\s*("?)(.*)\1/) |
| 136 |
{ |
{ |
| 137 |
# Ignore the NAME section |
# Ignore the NAME section |
| 139 |
{ |
{ |
| 140 |
<STDIN>; |
<STDIN>; |
| 141 |
next; |
next; |
| 142 |
} |
} |
| 143 |
|
|
| 144 |
&end_para(); |
&end_para(); |
| 145 |
my($title) = &do_line($2); |
my($title) = &do_line($2); |
| 146 |
if ($toc) |
if ($toc) |
| 147 |
{ |
{ |
| 148 |
printf("<li><a name=\"TOC%d\" href=\"#SEC%d\">$title</a>\n", |
printf("<li><a name=\"TOC%d\" href=\"#SEC%d\">$title</a>\n", |
| 149 |
$ref, $ref); |
$ref, $ref); |
| 150 |
printf TEMP ("<br><a name=\"SEC%d\" href=\"#TOC1\">$title</a><br>\n", |
printf TEMP ("<br><a name=\"SEC%d\" href=\"#TOC1\">$title</a><br>\n", |
| 151 |
$ref, $ref); |
$ref, $ref); |
| 152 |
$ref++; |
$ref++; |
| 153 |
} |
} |
| 154 |
else |
else |
| 155 |
{ |
{ |
| 156 |
print TEMP "<br><b>\n$title\n</b><br>\n"; |
print TEMP "<br><b>\n$title\n</b><br>\n"; |
| 157 |
} |
} |
| 158 |
} |
} |
| 159 |
elsif (/^\.SS\s*("?)(.*)\1/) |
elsif (/^\.SS\s*("?)(.*)\1/) |
| 160 |
{ |
{ |
| 161 |
&end_para(); |
&end_para(); |
| 162 |
my($title) = &do_line($2); |
my($title) = &do_line($2); |
| 163 |
print TEMP "<br><b>\n$title\n</b><br>\n"; |
print TEMP "<br><b>\n$title\n</b><br>\n"; |
| 164 |
} |
} |
| 165 |
elsif (/^\.B\s*(.*)/) |
elsif (/^\.B\s*(.*)/) |
| 166 |
{ |
{ |
| 167 |
&new_para() if (!$inpara); |
&new_para() if (!$inpara); |
| 168 |
$_ = &do_line($1); |
$_ = &do_line($1); |
| 169 |
s/"(.*?)"/$1/g; |
s/"(.*?)"/$1/g; |
| 170 |
print TEMP "<b>$_</b>\n"; |
print TEMP "<b>$_</b>\n"; |
| 171 |
$wrotetext = 1; |
$wrotetext = 1; |
| 172 |
} |
} |
| 173 |
elsif (/^\.I\s*(.*)/) |
elsif (/^\.I\s*(.*)/) |
| 174 |
{ |
{ |
| 175 |
&new_para() if (!$inpara); |
&new_para() if (!$inpara); |
| 176 |
$_ = &do_line($1); |
$_ = &do_line($1); |
| 177 |
s/"(.*?)"/$1/g; |
s/"(.*?)"/$1/g; |
| 178 |
print TEMP "<i>$_</i>\n"; |
print TEMP "<i>$_</i>\n"; |
| 179 |
$wrotetext = 1; |
$wrotetext = 1; |
| 180 |
} |
} |
| 181 |
|
|
| 182 |
# A comment that starts "HREF" takes the next line as a name that |
# A comment that starts "HREF" takes the next line as a name that |
| 183 |
# is turned into a hyperlink, using the text given, which might be |
# is turned into a hyperlink, using the text given, which might be |
| 184 |
# in a special font. If it ends in () or (digits) or punctuation, they |
# in a special font. If it ends in () or (digits) or punctuation, they |
| 185 |
# aren't part of the link. |
# aren't part of the link. |
| 186 |
|
|
| 187 |
elsif (/^\.\\"\s*HREF/) |
elsif (/^\.\\"\s*HREF/) |
| 188 |
{ |
{ |
| 189 |
$_=<STDIN>; |
$_=<STDIN>; |
| 190 |
chomp; |
chomp; |
| 191 |
$_ = &do_line($_); |
$_ = &do_line($_); |
| 192 |
$_ =~ s/\s+$//; |
$_ =~ s/\s+$//; |
| 193 |
$_ =~ /^(?:<.>)?([^<(]+)(?:\(\))?(?:<\/.>)?(?:\(\d+\))?[.,;:]?$/; |
$_ =~ /^(?:<.>)?([^<(]+)(?:\(\))?(?:<\/.>)?(?:\(\d+\))?[.,;:]?$/; |
| 194 |
print TEMP "<a href=\"$1.html\">$_</a>\n"; |
print TEMP "<a href=\"$1.html\">$_</a>\n"; |
| 195 |
} |
} |
| 196 |
|
|
| 197 |
# A comment that starts "HTML" inserts literal HTML |
# A comment that starts "HTML" inserts literal HTML |
| 198 |
|
|
| 199 |
elsif (/^\.\\"\s*HTML\s*(.*)/) |
elsif (/^\.\\"\s*HTML\s*(.*)/) |
| 200 |
{ |
{ |
| 201 |
print TEMP $1; |
print TEMP $1; |
| 202 |
} |
} |
| 203 |
|
|
| 204 |
# A comment that starts < inserts that HTML at the end of the |
# A comment that starts < inserts that HTML at the end of the |
| 205 |
# *next* input line - so as not to get a newline between them. |
# *next* input line - so as not to get a newline between them. |
| 206 |
|
|
| 207 |
elsif (/^\.\\"\s*(<.*>)/) |
elsif (/^\.\\"\s*(<.*>)/) |
| 208 |
{ |
{ |
| 209 |
my($markup) = $1; |
my($markup) = $1; |
| 210 |
$_=<STDIN>; |
$_=<STDIN>; |
| 211 |
chomp; |
chomp; |
| 212 |
$_ = &do_line($_); |
$_ = &do_line($_); |
| 213 |
$_ =~ s/\s+$//; |
$_ =~ s/\s+$//; |
| 214 |
print TEMP "$_$markup\n"; |
print TEMP "$_$markup\n"; |
| 215 |
} |
} |
| 216 |
|
|
| 217 |
# A comment that starts JOIN joins the next two lines together, with one |
# A comment that starts JOIN joins the next two lines together, with one |
| 218 |
# space between them. Then that line is processed. This is used in some |
# space between them. Then that line is processed. This is used in some |
| 219 |
# displays where two lines are needed for the "man" version. JOINSH works |
# displays where two lines are needed for the "man" version. JOINSH works |
| 220 |
# the same, except that it assumes this is a shell command, so removes |
# the same, except that it assumes this is a shell command, so removes |
| 221 |
# continuation backslashes. |
# continuation backslashes. |
| 222 |
|
|
| 223 |
elsif (/^\.\\"\s*JOIN(SH)?/) |
elsif (/^\.\\"\s*JOIN(SH)?/) |
| 224 |
{ |
{ |
| 225 |
my($one,$two); |
my($one,$two); |
| 230 |
$two =~ s/^\s+//; |
$two =~ s/^\s+//; |
| 231 |
$_ = "$one $two"; |
$_ = "$one $two"; |
| 232 |
redo; # Process the joined lines |
redo; # Process the joined lines |
| 233 |
} |
} |
| 234 |
|
|
| 235 |
# Ignore anything not recognized |
# Ignore anything not recognized |
| 236 |
|
|
| 237 |
next; |
next; |
| 238 |
} |
} |
| 239 |
|
|
| 240 |
# Line does not begin with a dot. Replace blank lines with new paragraphs |
# Line does not begin with a dot. Replace blank lines with new paragraphs |
| 241 |
|
|
| 242 |
if (/^\s*$/) |
if (/^\s*$/) |
| 243 |
{ |
{ |
| 244 |
&end_para() if ($wrotetext); |
&end_para() if ($wrotetext); |
| 245 |
next; |
next; |
| 246 |
} |
} |
| 247 |
|
|
| 248 |
# Convert fonts changes and output an ordinary line. Ensure that indented |
# Convert fonts changes and output an ordinary line. Ensure that indented |
| 249 |
# lines are marked as literal. |
# lines are marked as literal. |
| 250 |
|
|
| 251 |
$_ = &do_line($_); |
$_ = &do_line($_); |
| 252 |
&new_para() if (!$inpara); |
&new_para() if (!$inpara); |
| 253 |
|
|
| 254 |
if (/^\s/) |
if (/^\s/) |
| 255 |
{ |
{ |
| 256 |
if (!$inpre) |
if (!$inpre) |
| 257 |
{ |
{ |
| 258 |
print TEMP "<pre>\n"; |
print TEMP "<pre>\n"; |
| 259 |
$inpre = 1; |
$inpre = 1; |
| 260 |
} |
} |
| 261 |
} |
} |
| 262 |
elsif ($inpre) |
elsif ($inpre) |
| 263 |
{ |
{ |
| 264 |
print TEMP "</pre>\n"; |
print TEMP "</pre>\n"; |
| 265 |
$inpre = 0; |
$inpre = 0; |
| 266 |
} |
} |
| 267 |
|
|
| 268 |
print TEMP; |
# Add <br> to the end of a non-literal line if we are within .nf/.fi |
| 269 |
$wrotetext = 1; |
|
| 270 |
} |
$_ .= "<br>\n" if (!$inpre && $innf); |
| 271 |
|
|
| 272 |
|
print TEMP; |
| 273 |
|
$wrotetext = 1; |
| 274 |
|
} |
| 275 |
|
|
| 276 |
# The TOC, if present, will have been written - terminate it |
# The TOC, if present, will have been written - terminate it |
| 277 |
|
|
| 278 |
print "</ul>\n" if ($toc); |
print "</ul>\n" if ($toc); |
| 281 |
|
|
| 282 |
close(TEMP); |
close(TEMP); |
| 283 |
open(TEMP, "/tmp/$$") || die "Can't open /tmp/$$ for input\n"; |
open(TEMP, "/tmp/$$") || die "Can't open /tmp/$$ for input\n"; |
| 284 |
|
|
| 285 |
print while (<TEMP>); |
print while (<TEMP>); |
| 286 |
|
|
| 287 |
print <<End ; |
print <<End ; |