coreutils/man/help2man
2004-04-20 10:06:19 +00:00

628 lines
14 KiB
Perl
Executable File

#!/usr/bin/perl -w
# Generate a short man page from --help and --version output.
# Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003 Free Software
# Foundation, Inc.
# This program is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2, or (at your option)
# any later version.
# This program is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
# You should have received a copy of the GNU General Public License
# along with this program; if not, write to the Free Software Foundation,
# Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
# Written by Brendan O'Dea <bod@debian.org>
# Available from ftp://ftp.gnu.org/gnu/help2man/
use 5.005;
use strict;
use Getopt::Long;
use Text::Tabs qw(expand);
use POSIX qw(strftime setlocale LC_ALL);
use locale;
my $this_program = 'help2man';
my $this_version = '1.33';
my $have_gettext;
BEGIN {
eval {
require Locale::gettext;
Locale::gettext->import;
$have_gettext = 1;
};
unless ($have_gettext)
{
*gettext = sub { $_[0] };
*textdomain = sub {};
}
}
sub _ { gettext @_ }
sub N_ { $_[0] }
textdomain $this_program;
{
my ($user_locale) = grep defined && length,
(map $ENV{$_}, qw(LANGUAGE LC_ALL LC_MESSAGES LANG)), 'C';
sub kark # die with message formatted in the invoking user's locale
{
setlocale LC_ALL, $user_locale;
my $fmt = gettext shift;
die +(sprintf $fmt, @_), "\n";
}
}
my $version_info = sprintf _(<<'EOT'), $this_program, $this_version;
GNU %s %s
Copyright (C) 1997, 1998, 1999, 2000, 2001, 2002, 2003 Free Software
Foundation, Inc.
This is free software; see the source for copying conditions. There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
Written by Brendan O'Dea <bod@debian.org>
EOT
my $help_info = sprintf _(<<'EOT'), $this_program, $this_program;
`%s' generates a man page out of `--help' and `--version' output.
Usage: %s [OPTION]... EXECUTABLE
--program_name=NAME override any name found in output of
`EXECUTABLE --version'
-n, --name=STRING description for the NAME paragraph
-s, --section=SECTION section number for manual page (1, 6, 8)
-m, --manual=TEXT name of manual (User Commands, ...)
-S, --source=TEXT source of program (FSF, Debian, ...)
-L, --locale=STRING select locale (default "C")
-i, --include=FILE include material from `FILE'
-I, --opt-include=FILE include material from `FILE' if it exists
-o, --output=FILE send output to `FILE'
-p, --info-page=TEXT name of Texinfo manual
-N, --no-info suppress pointer to Texinfo manual
--help print this help, then exit
--version print version number, then exit
EXECUTABLE should accept `--help' and `--version' options although
alternatives may be specified using:
-h, --help-option=STRING help option string
-v, --version-option=STRING version option string
Report bugs to <bug-help2man@gnu.org>.
EOT
my $section = 1;
my $manual = '';
my $source = '';
my $locale = 'C';
my $help_option = '--help';
my $version_option = '--version';
my ($opt_name, @opt_include, $opt_output, $opt_info, $opt_no_info);
my $opt_program_name;
my %opt_def = (
'program-name=s' => \$opt_program_name,
'n|name=s' => \$opt_name,
's|section=s' => \$section,
'm|manual=s' => \$manual,
'S|source=s' => \$source,
'L|locale=s' => \$locale,
'i|include=s' => sub { push @opt_include, [ pop, 1 ] },
'I|opt-include=s' => sub { push @opt_include, [ pop, 0 ] },
'o|output=s' => \$opt_output,
'p|info-page=s' => \$opt_info,
'N|no-info' => \$opt_no_info,
'h|help-option=s' => \$help_option,
'v|version-option=s' => \$version_option,
);
# Parse options.
Getopt::Long::config('bundling');
GetOptions (%opt_def,
help => sub { print $help_info; exit },
version => sub { print $version_info; exit },
) or die $help_info;
die $help_info unless @ARGV == 1;
die "$this_program: no locale support (Locale::gettext required)\n"
unless $locale eq 'C' or $have_gettext;
# Set localisation of date and executable's ouput.
delete @ENV{qw(LANGUAGE LC_MESSAGES LANG)};
setlocale LC_ALL, $ENV{LC_ALL} = $locale;
my %include = ();
my %append = ();
my @include = (); # retain order given in include file
# Process include file (if given). Format is:
#
# [section name]
# verbatim text
#
# or
#
# /pattern/
# verbatim text
#
while (@opt_include)
{
my ($inc, $required) = @{shift @opt_include};
next unless -f $inc or $required;
kark N_("%s: can't open `%s' (%s)"), $this_program, $inc, $!
unless open INC, $inc;
my $key;
my $hash = \%include;
while (<INC>)
{
# [section]
if (/^\[([^]]+)\]/)
{
$key = uc $1;
$key =~ s/^\s+//;
$key =~ s/\s+$//;
$hash = \%include;
push @include, $key unless $include{$key};
next;
}
# /pattern/
if (m!^/(.*)/([ims]*)!)
{
my $pat = $2 ? "(?$2)$1" : $1;
# Check pattern.
eval { $key = qr($pat) };
if ($@)
{
$@ =~ s/ at .*? line \d.*//;
die "$inc:$.:$@";
}
$hash = \%append;
next;
}
# Check for options before the first section--anything else is
# silently ignored, allowing the first for comments and
# revision info.
unless ($key)
{
# handle options
if (/^-/)
{
local @ARGV = split;
GetOptions %opt_def;
}
next;
}
$hash->{$key} ||= '';
$hash->{$key} .= $_;
}
close INC;
kark N_("%s: no valid information found in `%s'"), $this_program, $inc
unless $key;
}
# Compress trailing blank lines.
for my $hash (\(%include, %append))
{
for (keys %$hash) { $hash->{$_} =~ s/\n+$/\n/ }
}
# Grab help and version info from executable.
my ($help_text, $version_text) = map {
join '', map { s/ +$//; expand $_ } `$ARGV[0] $_ 2>/dev/null`
or kark N_("%s: can't get `%s' info from %s"), $this_program,
$_, $ARGV[0]
} $help_option, $version_option;
my $date = strftime "%B %Y", localtime;
(my $program = $ARGV[0]) =~ s!.*/!!;
my $package = $program;
my $version;
if ($opt_output)
{
unlink $opt_output or kark N_("%s: can't unlink %s (%s)"),
$this_program, $opt_output, $! if -e $opt_output;
open STDOUT, ">$opt_output"
or kark N_("%s: can't create %s (%s)"), $this_program, $opt_output, $!;
}
# The first line of the --version information is assumed to be in one
# of the following formats:
#
# <version>
# <program> <version>
# {GNU,Free} <program> <version>
# <program> ({GNU,Free} <package>) <version>
# <program> - {GNU,Free} <package> <version>
#
# and seperated from any copyright/author details by a blank line.
($_, $version_text) = split /\n+/, $version_text, 2;
if (/^(\S+) +\(((?:GNU|Free) +[^)]+)\) +(.*)/ or
/^(\S+) +- *((?:GNU|Free) +\S+) +(.*)/)
{
$program = $1;
$package = $2;
$version = $3;
}
elsif (/^((?:GNU|Free) +)?(\S+) +(.*)/)
{
$program = $2;
$package = $1 ? "$1$2" : $2;
$version = $3;
}
else
{
$version = $_;
}
$program =~ s!.*/!!;
$opt_program_name
and $program = $opt_program_name;
# No info for `info' itself.
$opt_no_info = 1 if $program eq 'info';
# --name overrides --include contents.
$include{_('NAME')} = "$program \\- $opt_name\n" if $opt_name;
# Default (useless) NAME paragraph.
$include{_('NAME')} ||= sprintf _("%s \\- manual page for %s %s") . "\n",
$program, $program, $version;
# Man pages traditionally have the page title in caps.
my $PROGRAM = uc $program;
# Set default page head/footers
$source ||= "$program $version";
unless ($manual)
{
for ($section)
{
if (/^(1[Mm]|8)/) { $manual = _('System Administration Utilities') }
elsif (/^6/) { $manual = _('Games') }
else { $manual = _('User Commands') }
}
}
# Extract usage clause(s) [if any] for SYNOPSIS.
my $PAT_USAGE = _('Usage');
my $PAT_USAGE_CONT = _('or');
if ($help_text =~ s/^($PAT_USAGE):( +(\S+))(.*)((?:\n(?: {6}\1| *($PAT_USAGE_CONT): +\S).*)*)//om)
{
my @syn = $3 . $4;
if ($_ = $5)
{
s/^\n//;
for (split /\n/) { s/^ *(($PAT_USAGE_CONT): +)?//o; push @syn, $_ }
}
my $synopsis = '';
for (@syn)
{
$synopsis .= ".br\n" if $synopsis;
s!^\S*/!!;
s/^(\S+) *//;
$synopsis .= ".B $1\n";
s/\s+$//;
s/(([][]|\.\.+)+)/\\fR$1\\fI/g;
s/^/\\fI/ unless s/^\\fR//;
$_ .= '\fR';
s/(\\fI)( *)/$2$1/g;
s/\\fI\\fR//g;
s/^\\fR//;
s/\\fI$//;
s/^\./\\&./;
$synopsis .= "$_\n";
}
$include{_('SYNOPSIS')} ||= $synopsis;
}
# Process text, initial section is DESCRIPTION.
my $sect = _('DESCRIPTION');
$_ = "$help_text\n\n$version_text";
# Normalise paragraph breaks.
s/^\n+//;
s/\n*$/\n/;
s/\n\n+/\n\n/g;
# Temporarily exchange leading dots, apostrophes and backslashes for
# tokens.
s/^\./\x80/mg;
s/^'/\x81/mg;
s/\\/\x82/g;
my $PAT_BUGS = _('Report +bugs|Email +bug +reports +to');
my $PAT_AUTHOR = _('Written +by');
my $PAT_OPTIONS = _('Options');
my $PAT_EXAMPLES = _('Examples');
my $PAT_FREE_SOFTWARE = _('This +is +free +software');
# Start a new paragraph (if required) for these.
s/([^\n])\n($PAT_BUGS|$PAT_AUTHOR)/$1\n\n$2/og;
sub convert_option;
while (length)
{
# Convert some standard paragraph names.
if (s/^($PAT_OPTIONS): *\n//o)
{
$sect = _('OPTIONS');
next;
}
elsif (s/^($PAT_EXAMPLES): *\n//o)
{
$sect = _('EXAMPLES');
next;
}
# Copyright section
if (/^Copyright +[(\xa9]/)
{
$sect = _('COPYRIGHT');
$include{$sect} ||= '';
$include{$sect} .= ".PP\n" if $include{$sect};
my $copy;
($copy, $_) = split /\n\n/, $_, 2;
for ($copy)
{
# Add back newline
s/\n*$/\n/;
# Convert iso9959-1 copyright symbol or (c) to nroff
# character.
s/^Copyright +(?:\xa9|\([Cc]\))/Copyright \\(co/mg;
# Insert line breaks before additional copyright messages
# and the disclaimer.
s/(.)\n(Copyright |$PAT_FREE_SOFTWARE)/$1\n.br\n$2/og;
# Join hyphenated lines.
s/([A-Za-z])-\n */$1/g;
}
$include{$sect} .= $copy;
$_ ||= '';
next;
}
# Catch bug report text.
if (/^($PAT_BUGS) /o)
{
$sect = _('REPORTING BUGS');
}
# Author section.
elsif (/^($PAT_AUTHOR)/o)
{
$sect = _('AUTHOR');
}
# Examples, indicated by an indented leading $, % or > are
# rendered in a constant width font.
if (/^( +)([\$\%>] )\S/)
{
my $indent = $1;
my $prefix = $2;
my $break = '.IP';
$include{$sect} ||= '';
while (s/^$indent\Q$prefix\E(\S.*)\n*//)
{
$include{$sect} .= "$break\n\\f(CW$prefix$1\\fR\n";
$break = '.br';
}
next;
}
my $matched = '';
$include{$sect} ||= '';
# Sub-sections have a trailing colon and the second line indented.
if (s/^(\S.*:) *\n / /)
{
$matched .= $& if %append;
$include{$sect} .= qq(.SS "$1"\n);
}
my $indent = 0;
my $content = '';
# Option with description.
if (s/^( {1,10}([+-]\S.*?))(?:( +(?!-))|\n( {20,}))(\S.*)\n//)
{
$matched .= $& if %append;
$indent = length ($4 || "$1$3");
$content = ".TP\n\x83$2\n\x83$5\n";
unless ($4)
{
# Indent may be different on second line.
$indent = length $& if /^ {20,}/;
}
}
# Option without description.
elsif (s/^ {1,10}([+-]\S.*)\n//)
{
$matched .= $& if %append;
$content = ".HP\n\x83$1\n";
$indent = 80; # not continued
}
# Indented paragraph with tag.
elsif (s/^( +(\S.*?) +)(\S.*)\n//)
{
$matched .= $& if %append;
$indent = length $1;
$content = ".TP\n\x83$2\n\x83$3\n";
}
# Indented paragraph.
elsif (s/^( +)(\S.*)\n//)
{
$matched .= $& if %append;
$indent = length $1;
$content = ".IP\n\x83$2\n";
}
# Left justified paragraph.
else
{
s/(.*)\n//;
$matched .= $& if %append;
$content = ".PP\n" if $include{$sect};
$content .= "$1\n";
}
# Append continuations.
while ($indent ? s/^ {$indent}(\S.*)\n// : s/^(\S.*)\n//)
{
$matched .= $& if %append;
$content .= "\x83$1\n"
}
# Move to next paragraph.
s/^\n+//;
for ($content)
{
# Leading dot and apostrophe protection.
s/\x83\./\x80/g;
s/\x83'/\x81/g;
s/\x83//g;
# Convert options.
s/(^| )(-[][\w=-]+)/$1 . convert_option $2/mge;
}
# Check if matched paragraph contains /pat/.
if (%append)
{
for my $pat (keys %append)
{
if ($matched =~ $pat)
{
$content .= ".PP\n" unless $append{$pat} =~ /^\./;
$content .= $append{$pat};
}
}
}
$include{$sect} .= $content;
}
# Refer to the real documentation.
unless ($opt_no_info)
{
my $info_page = $opt_info || $program;
$sect = _('SEE ALSO');
$include{$sect} ||= '';
$include{$sect} .= ".PP\n" if $include{$sect};
$include{$sect} .= sprintf _(<<'EOT'), $program, $program, $info_page;
The full documentation for
.B %s
is maintained as a Texinfo manual. If the
.B info
and
.B %s
programs are properly installed at your site, the command
.IP
.B info %s
.PP
should give you access to the complete manual.
EOT
}
# Output header.
print <<EOT;
.\\" DO NOT MODIFY THIS FILE! It was generated by $this_program $this_version.
.TH $PROGRAM "$section" "$date" "$source" "$manual"
EOT
# Section ordering.
my @pre = (_('NAME'), _('SYNOPSIS'), _('DESCRIPTION'), _('OPTIONS'),
_('EXAMPLES'));
my @post = (_('AUTHOR'), _('REPORTING BUGS'), _('COPYRIGHT'), _('SEE ALSO'));
my $filter = join '|', @pre, @post;
# Output content.
for my $sect (@pre, (grep ! /^($filter)$/o, @include), @post)
{
if ($include{$sect})
{
my $lsect = gettext $sect;
my $quote = $lsect =~ /\W/ ? '"' : '';
print ".SH $quote$lsect$quote\n";
for ($include{$sect})
{
# Replace leading dot, apostrophe and backslash tokens.
s/\x80/\\&./g;
s/\x81/\\&'/g;
s/\x82/\\e/g;
# Convert some latin1 chars to troff equivalents
s/\xa0/\\ /g; # non-breaking space
print;
}
}
}
close STDOUT or kark N_("%s: error writing to %s (%s)"), $this_program,
$opt_output || 'stdout', $!;
exit;
# Convert option dashes to \- to stop nroff from hyphenating 'em, and
# embolden. Option arguments get italicised.
sub convert_option
{
local $_ = '\fB' . shift;
s/-/\\-/g;
unless (s/\[=(.*)\]$/\\fR[=\\fI$1\\fR]/)
{
s/=(.)/\\fR=\\fI$1/;
s/ (.)/ \\fI$1/;
$_ .= '\fR';
}
$_;
}