1#!/usr/bin/env perl
2
3# Generate a short man page from --help and --version output.
4# Copyright (C) 1997-2020 Free Software Foundation, Inc.
5
6# This program is free software; you can redistribute it and/or modify
7# it under the terms of the GNU General Public License as published by
8# the Free Software Foundation; either version 2, or (at your option)
9# any later version.
10
11# This program is distributed in the hope that it will be useful,
12# but WITHOUT ANY WARRANTY; without even the implied warranty of
13# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
14# GNU General Public License for more details.
15
16# You should have received a copy of the GNU General Public License
17# along with this program; if not, write to the Free Software Foundation,
18# Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
19
20# Written by Brendan O'Dea <bod@debian.org>
21# Available from ftp://ftp.gnu.org/gnu/help2man/
22
23use 5.005;
24use strict;
25use Getopt::Long;
26use Text::Tabs qw(expand);
27use POSIX qw(strftime setlocale LC_TIME);
28
29my $this_program = 'help2man';
30my $this_version = '1.28';
31my $version_info = <<EOT;
32GNU $this_program $this_version
33
34Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002 Free Software Foundation, Inc.
35This is free software; see the source for copying conditions.  There is NO
36warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
37
38Written by Brendan O'Dea <bod\@debian.org>
39EOT
40
41my $help_info = <<EOT;
42`$this_program' generates a man page out of `--help' and `--version' output.
43
44Usage: $this_program [OPTION]... EXECUTABLE
45
46 -n, --name=STRING       description for the NAME paragraph
47 -s, --section=SECTION   section number for manual page (1, 6, 8)
48 -m, --manual=TEXT       name of manual (User Commands, ...)
49 -S, --source=TEXT       source of program (FSF, Debian, ...)
50 -i, --include=FILE      include material from `FILE'
51 -I, --opt-include=FILE  include material from `FILE' if it exists
52 -o, --output=FILE       send output to `FILE'
53 -p, --info-page=TEXT    name of Texinfo manual
54 -N, --no-info           suppress pointer to Texinfo manual
55     --help              print this help, then exit
56     --version           print version number, then exit
57
58EXECUTABLE should accept `--help' and `--version' options although
59alternatives may be specified using:
60
61 -h, --help-option=STRING     help option string
62 -v, --version-option=STRING  version option string
63
64Report bugs to <bug-help2man\@gnu.org>.
65EOT
66
67my $section = 1;
68my $manual = '';
69my $source = '';
70my $help_option = '--help';
71my $version_option = '--version';
72my ($opt_name, @opt_include, $opt_output, $opt_info, $opt_no_info);
73
74my %opt_def = (
75    'n|name=s'		 => \$opt_name,
76    's|section=s'	 => \$section,
77    'm|manual=s'	 => \$manual,
78    'S|source=s'	 => \$source,
79    'i|include=s'	 => sub { push @opt_include, [ pop, 1 ] },
80    'I|opt-include=s'	 => sub { push @opt_include, [ pop, 0 ] },
81    'o|output=s'	 => \$opt_output,
82    'p|info-page=s'	 => \$opt_info,
83    'N|no-info'		 => \$opt_no_info,
84    'h|help-option=s'	 => \$help_option,
85    'v|version-option=s' => \$version_option,
86);
87
88# Parse options.
89Getopt::Long::config('bundling');
90GetOptions (%opt_def,
91    help    => sub { print $help_info; exit },
92    version => sub { print $version_info; exit },
93) or die $help_info;
94
95die $help_info unless @ARGV == 1;
96
97my %include = ();
98my %append = ();
99my @include = (); # retain order given in include file
100
101# Process include file (if given).  Format is:
102#
103#   [section name]
104#   verbatim text
105#
106# or
107#
108#   /pattern/
109#   verbatim text
110#
111
112while (@opt_include)
113{
114    my ($inc, $required) = @{shift @opt_include};
115
116    next unless -f $inc or $required;
117    die "$this_program: can't open `$inc' ($!)\n"
118	unless open INC, $inc;
119
120    my $key;
121    my $hash = \%include;
122
123    while (<INC>)
124    {
125	# [section]
126	if (/^\[([^]]+)\]/)
127	{
128	    $key = uc $1;
129	    $key =~ s/^\s+//;
130	    $key =~ s/\s+$//;
131	    $hash = \%include;
132	    push @include, $key unless $include{$key};
133	    next;
134	}
135
136	# /pattern/
137	if (m!^/(.*)/([ims]*)!)
138	{
139	    my $pat = $2 ? "(?$2)$1" : $1;
140
141	    # Check pattern.
142	    eval { $key = qr($pat) };
143	    if ($@)
144	    {
145		$@ =~ s/ at .*? line \d.*//;
146		die "$inc:$.:$@";
147	    }
148
149	    $hash = \%append;
150	    next;
151	}
152
153	# Check for options before the first section--anything else is
154	# silently ignored, allowing the first for comments and
155	# revision info.
156	unless ($key)
157	{
158	    # handle options
159	    if (/^-/)
160	    {
161		local @ARGV = split;
162		GetOptions %opt_def;
163	    }
164
165	    next;
166	}
167
168	$hash->{$key} ||= '';
169	$hash->{$key} .= $_;
170    }
171
172    close INC;
173
174    die "$this_program: no valid information found in `$inc'\n"
175	unless $key;
176}
177
178# Compress trailing blank lines.
179for my $hash (\(%include, %append))
180{
181    for (keys %$hash) { $hash->{$_} =~ s/\n+$/\n/ }
182}
183
184# Turn off localisation of executable's ouput.
185@ENV{qw(LANGUAGE LANG LC_ALL)} = ('C') x 3;
186
187# Turn off localisation of date (for strftime).
188setlocale LC_TIME, 'C';
189
190# Grab help and version info from executable.
191my ($help_text, $version_text) = map {
192    join '', map { s/ +$//; expand $_ } `$ARGV[0] $_ 2>/dev/null`
193	or die "$this_program: can't get `$_' info from $ARGV[0]\n"
194} $help_option, $version_option;
195
196my $date = strftime "%B %Y", localtime;
197(my $program = $ARGV[0]) =~ s!.*/!!;
198my $package = $program;
199my $version;
200
201if ($opt_output)
202{
203    unlink $opt_output
204	or die "$this_program: can't unlink $opt_output ($!)\n"
205	if -e $opt_output;
206
207    open STDOUT, ">$opt_output"
208	or die "$this_program: can't create $opt_output ($!)\n";
209}
210
211# The first line of the --version information is assumed to be in one
212# of the following formats:
213#
214#   <version>
215#   <program> <version>
216#   {GNU,Free} <program> <version>
217#   <program> ({GNU,Free} <package>) <version>
218#   <program> - {GNU,Free} <package> <version>
219#
220# and seperated from any copyright/author details by a blank line.
221
222($_, $version_text) = split /\n+/, $version_text, 2;
223
224if (/^(\S+) +\(((?:GNU|Free) +[^)]+)\) +(.*)/ or
225    /^(\S+) +- *((?:GNU|Free) +\S+) +(.*)/)
226{
227    $program = $1;
228    $package = $2;
229    $version = $3;
230}
231elsif (/^((?:GNU|Free) +)?(\S+) +(.*)/)
232{
233    $program = $2;
234    $package = $1 ? "$1$2" : $2;
235    $version = $3;
236}
237else
238{
239    $version = $_;
240}
241
242$program =~ s!.*/!!;
243
244# No info for `info' itself.
245$opt_no_info = 1 if $program eq 'info';
246
247# --name overrides --include contents.
248$include{NAME} = "$program \\- $opt_name\n" if $opt_name;
249
250# Default (useless) NAME paragraph.
251$include{NAME} ||= "$program \\- manual page for $program $version\n";
252
253# Man pages traditionally have the page title in caps.
254my $PROGRAM = uc $program;
255
256# Set default page head/footers
257$source ||= "$program $version";
258unless ($manual)
259{
260    for ($section)
261    {
262	if (/^(1[Mm]|8)/) { $manual = 'System Administration Utilities' }
263	elsif (/^6/)	  { $manual = 'Games' }
264	else		  { $manual = 'User Commands' }
265    }
266}
267
268# Extract usage clause(s) [if any] for SYNOPSIS.
269if ($help_text =~ s/^Usage:( +(\S+))(.*)((?:\n(?: {6}\1| *or: +\S).*)*)//m)
270{
271    my @syn = $2 . $3;
272
273    if ($_ = $4)
274    {
275	s/^\n//;
276	for (split /\n/) { s/^ *(or: +)?//; push @syn, $_ }
277    }
278
279    my $synopsis = '';
280    for (@syn)
281    {
282	$synopsis .= ".br\n" if $synopsis;
283	s!^\S*/!!;
284	s/^(\S+) *//;
285	$synopsis .= ".B $1\n";
286	s/\s+$//;
287	s/(([][]|\.\.+)+)/\\fR$1\\fI/g;
288	s/^/\\fI/ unless s/^\\fR//;
289	$_ .= '\fR';
290	s/(\\fI)( *)/$2$1/g;
291	s/\\fI\\fR//g;
292	s/^\\fR//;
293	s/\\fI$//;
294	s/^\./\\&./;
295
296	$synopsis .= "$_\n";
297    }
298
299    $include{SYNOPSIS} ||= $synopsis;
300}
301
302# Process text, initial section is DESCRIPTION.
303my $sect = 'DESCRIPTION';
304$_ = "$help_text\n\n$version_text";
305
306# Normalise paragraph breaks.
307s/^\n+//;
308s/\n*$/\n/;
309s/\n\n+/\n\n/g;
310
311# Temporarily exchange leading dots, apostrophes and backslashes for
312# tokens.
313s/^\./\x80/mg;
314s/^'/\x81/mg;
315s/\\/\x82/g;
316
317# Start a new paragraph (if required) for these.
318s/([^\n])\n(Report +bugs|Email +bug +reports +to|Written +by)/$1\n\n$2/g;
319
320sub convert_option;
321
322while (length)
323{
324    # Convert some standard paragraph names.
325    if (s/^(Options|Examples): *\n//)
326    {
327	$sect = uc $1;
328	next;
329    }
330
331    # Copyright section
332    if (/^Copyright +[(\xa9]/)
333    {
334	$sect = 'COPYRIGHT';
335	$include{$sect} ||= '';
336	$include{$sect} .= ".PP\n" if $include{$sect};
337
338	my $copy;
339	($copy, $_) = split /\n\n/, $_, 2;
340
341	for ($copy)
342	{
343	    # Add back newline
344	    s/\n*$/\n/;
345
346	    # Convert iso9959-1 copyright symbol or (c) to nroff
347	    # character.
348	    s/^Copyright +(?:\xa9|\([Cc]\))/Copyright \\(co/mg;
349
350	    # Insert line breaks before additional copyright messages
351	    # and the disclaimer.
352	    s/(.)\n(Copyright |This +is +free +software)/$1\n.br\n$2/g;
353
354	    # Join hyphenated lines.
355	    s/([A-Za-z])-\n */$1/g;
356	}
357
358	$include{$sect} .= $copy;
359	$_ ||= '';
360	next;
361    }
362
363    # Catch bug report text.
364    if (/^(Report +bugs|Email +bug +reports +to) /)
365    {
366	$sect = 'REPORTING BUGS';
367    }
368
369    # Author section.
370    elsif (/^Written +by/)
371    {
372	$sect = 'AUTHOR';
373    }
374
375    # Examples, indicated by an indented leading $, % or > are
376    # rendered in a constant width font.
377    if (/^( +)([\$\%>] )\S/)
378    {
379	my $indent = $1;
380	my $prefix = $2;
381	my $break = '.IP';
382	$include{$sect} ||= '';
383	while (s/^$indent\Q$prefix\E(\S.*)\n*//)
384	{
385	    $include{$sect} .= "$break\n\\f(CW$prefix$1\\fR\n";
386	    $break = '.br';
387	}
388
389	next;
390    }
391
392    my $matched = '';
393    $include{$sect} ||= '';
394
395    # Sub-sections have a trailing colon and the second line indented.
396    if (s/^(\S.*:) *\n / /)
397    {
398	$matched .= $& if %append;
399	$include{$sect} .= qq(.SS "$1"\n);
400    }
401
402    my $indent = 0;
403    my $content = '';
404
405    # Option with description.
406    if (s/^( {1,10}([+-]\S.*?))(?:(  +(?!-))|\n( {20,}))(\S.*)\n//)
407    {
408	$matched .= $& if %append;
409	$indent = length ($4 || "$1$3");
410	$content = ".TP\n\x83$2\n\x83$5\n";
411	unless ($4)
412	{
413	    # Indent may be different on second line.
414	    $indent = length $& if /^ {20,}/;
415	}
416    }
417
418    # Option without description.
419    elsif (s/^ {1,10}([+-]\S.*)\n//)
420    {
421	$matched .= $& if %append;
422	$content = ".HP\n\x83$1\n";
423	$indent = 80; # not continued
424    }
425
426    # Indented paragraph with tag.
427    elsif (s/^( +(\S.*?)  +)(\S.*)\n//)
428    {
429	$matched .= $& if %append;
430	$indent = length $1;
431	$content = ".TP\n\x83$2\n\x83$3\n";
432    }
433
434    # Indented paragraph.
435    elsif (s/^( +)(\S.*)\n//)
436    {
437	$matched .= $& if %append;
438	$indent = length $1;
439	$content = ".IP\n\x83$2\n";
440    }
441
442    # Left justified paragraph.
443    else
444    {
445	s/(.*)\n//;
446	$matched .= $& if %append;
447	$content = ".PP\n" if $include{$sect};
448	$content .= "$1\n";
449    }
450
451    # Append continuations.
452    while (s/^ {$indent}(\S.*)\n//)
453    {
454	$matched .= $& if %append;
455	$content .= "\x83$1\n"
456    }
457
458    # Move to next paragraph.
459    s/^\n+//;
460
461    for ($content)
462    {
463	# Leading dot and apostrophe protection.
464	s/\x83\./\x80/g;
465	s/\x83'/\x81/g;
466	s/\x83//g;
467
468	# Convert options.
469	s/(^| )(-[][\w=-]+)/$1 . convert_option $2/mge;
470    }
471
472    # Check if matched paragraph contains /pat/.
473    if (%append)
474    {
475	for my $pat (keys %append)
476	{
477	    if ($matched =~ $pat)
478	    {
479		$content .= ".PP\n" unless $append{$pat} =~ /^\./;
480		$content .= $append{$pat};
481	    }
482	}
483    }
484
485    $include{$sect} .= $content;
486}
487
488# Refer to the real documentation.
489unless ($opt_no_info)
490{
491    my $info_page = $opt_info || $program;
492
493    $sect = 'SEE ALSO';
494    $include{$sect} ||= '';
495    $include{$sect} .= ".PP\n" if $include{$sect};
496    $include{$sect} .= <<EOT;
497The full documentation for
498.B $program
499is maintained as a Texinfo manual.  If the
500.B info
501and
502.B $program
503programs are properly installed at your site, the command
504.IP
505.B info $info_page
506.PP
507should give you access to the complete manual.
508EOT
509}
510
511# Output header.
512print <<EOT;
513.\\" DO NOT MODIFY THIS FILE!  It was generated by $this_program $this_version.
514.TH $PROGRAM "$section" "$date" "$source" "$manual"
515EOT
516
517# Section ordering.
518my @pre = qw(NAME SYNOPSIS DESCRIPTION OPTIONS EXAMPLES);
519my @post = ('AUTHOR', 'REPORTING BUGS', 'COPYRIGHT', 'SEE ALSO');
520my $filter = join '|', @pre, @post;
521
522# Output content.
523for (@pre, (grep ! /^($filter)$/o, @include), @post)
524{
525    if ($include{$_})
526    {
527	my $quote = /\W/ ? '"' : '';
528	print ".SH $quote$_$quote\n";
529
530	for ($include{$_})
531	{
532	    # Replace leading dot, apostrophe and backslash tokens.
533	    s/\x80/\\&./g;
534	    s/\x81/\\&'/g;
535	    s/\x82/\\e/g;
536	    print;
537	}
538    }
539}
540
541exit;
542
543# Convert option dashes to \- to stop nroff from hyphenating 'em, and
544# embolden.  Option arguments get italicised.
545sub convert_option
546{
547    local $_ = '\fB' . shift;
548
549    s/-/\\-/g;
550    unless (s/\[=(.*)\]$/\\fR[=\\fI$1\\fR]/)
551    {
552	s/=(.)/\\fR=\\fI$1/;
553	s/ (.)/ \\fI$1/;
554	$_ .= '\fR';
555    }
556
557    $_;
558}
559