|
|
@@ -1,335 +0,0 @@
|
|
|
-#!/bin/sh -- #perl, to stop looping
|
|
|
-eval 'exec $GRASS_PERL -S $0 ${1+"$@"}'
|
|
|
- if 0;
|
|
|
-
|
|
|
-############################################################################
|
|
|
-#
|
|
|
-# MODULE: g.html2man
|
|
|
-# AUTHOR(S): Daniel Calvelo Aros
|
|
|
-# PURPOSE: convert HTML formatted page to MAN page format
|
|
|
-# COPYRIGHT: (C) 2004 by the GRASS Development Team
|
|
|
-#
|
|
|
-# This program is free software under the GNU General Public
|
|
|
-# License (>=v2). Read the file COPYING that comes with GRASS
|
|
|
-# for details.
|
|
|
-#
|
|
|
-#############################################################################
|
|
|
-
|
|
|
-# Note: the GRASS_PERL variables must be set!
|
|
|
-# Usage:
|
|
|
-# perl g.html2man <html file> <man file> <section>
|
|
|
-
|
|
|
-our $preformat, $in_bu, $in_nb, @nb_n;
|
|
|
-our $suppress = 0;
|
|
|
-
|
|
|
-sub CvtTable {
|
|
|
-#
|
|
|
-# converts <table> structures to tbl commands.
|
|
|
-#
|
|
|
-# limitations: malformed (sloppy html) tables won't convert
|
|
|
-# properly, eg no closing tags are bad.
|
|
|
-# <th> headers are handled; nested <table>s are not
|
|
|
-#
|
|
|
-# since tables may be line-formatted randomly, this function
|
|
|
-# works on a full file; thus, the original line-oriented
|
|
|
-# unbuffered design had to be changed.
|
|
|
-#
|
|
|
-# modify $format if centered headings and left-aligned
|
|
|
-# contents don't suit you.
|
|
|
-#
|
|
|
-# also, groff recognizes the tbl commands and preprocesses
|
|
|
-# accordingly; for other roff formatters that is uncertain.
|
|
|
-# it might be advisable to put a '.\" t' line at the beginning
|
|
|
-# of the output file
|
|
|
-#
|
|
|
-# TODO: rewrite g.html2man using the HTML::Parser module
|
|
|
-#
|
|
|
- my $i = shift;
|
|
|
- return $i unless $i =~ m|<table.*?>(.+?)</table>|is;
|
|
|
- my $header = ".TS\nexpand box;\n";
|
|
|
- $_ = $1;
|
|
|
- ($one_row,) = m|<tr.*?>(.+?)</tr>|is ;
|
|
|
- $ncols = -1 + scalar split /<td/i, $one_row;
|
|
|
- $has_header = m|<th|i ;
|
|
|
- s|[\n\r]||g;
|
|
|
- s|</tr.*?>|\n.br\n|gis;
|
|
|
- s|</td.*?>|\t|gis;
|
|
|
- foreach $tag ( qw(<td.*?> <th.*?> </th> <tr.*?>) ){
|
|
|
- s/$tag//gi;
|
|
|
- }
|
|
|
- $format = $has_header
|
|
|
- ? "cb"x$ncols . "\n"."-"x$ncols . "\n"
|
|
|
- : "";
|
|
|
- $format .= "l"x$ncols . ".\n";
|
|
|
- $i =~ s|<table.*?>.+?<table>|$header$format$_.TE\n|is;
|
|
|
- return CvtTable($i);
|
|
|
-}
|
|
|
-
|
|
|
-sub basename {
|
|
|
- $slash = rindex($_[0], "/");
|
|
|
- $suffix = rindex($_[0], $_[1]);
|
|
|
- if ($suffix == -1) {$suffix=length($_[0])};
|
|
|
- return substr($_[0], $slash+1, $suffix-$slash-1);
|
|
|
-}
|
|
|
-
|
|
|
-sub GetTitle {
|
|
|
- while (<INP>) {
|
|
|
- if (m#<TITLE>.*MAN page for:#i) {
|
|
|
- s#<TITLE>##i;
|
|
|
- s#MAN page for:##i;
|
|
|
- s#</TITLE>##i;
|
|
|
- @TITLE = split;
|
|
|
- return 1;
|
|
|
- }
|
|
|
- if (m#<TITLE>#i) {
|
|
|
- return 1;
|
|
|
- }
|
|
|
- }
|
|
|
- return 0;
|
|
|
-}
|
|
|
-
|
|
|
-sub PrintHeader {
|
|
|
- print OUT ".TH $_[0] $_[1] \"\" \"GRASS " . $ENV{'VERSION_NUMBER'} . "\" \"Grass User's Manual\"\n";
|
|
|
-## print OUT ".so /usr/share/lib/tmac/sml\n";
|
|
|
-## print OUT ".so /usr/share/lib/tmac/rsml\n";
|
|
|
-}
|
|
|
-
|
|
|
-sub DoStrip {
|
|
|
- if (m#<.*TITLE>#i) { return 1};
|
|
|
- if (m/<.*HEADER>/i) { return 1};
|
|
|
- if (m/<.*HEAD>/i) { return 1};
|
|
|
- if (m/<.*BODY>/i) { return 1};
|
|
|
- if (m/<.*HEADER>/i) { return 1};
|
|
|
- if (m/<.*BLINK>/i) { return 1};
|
|
|
- if (m/^<A HREF="#toc/i) {return 1};
|
|
|
- if (s#<A HREF=".*">##i) {
|
|
|
- s#</A>##i;
|
|
|
- }
|
|
|
- return 0;
|
|
|
-}
|
|
|
-
|
|
|
-sub DoLine {
|
|
|
-
|
|
|
- if (m#</PRE>#i) {
|
|
|
- s#</PRE>#\n.DE\n\\fR#i;
|
|
|
- $preformat = 0;
|
|
|
- return 1;
|
|
|
- }
|
|
|
-### if ($preformat) {return 1};
|
|
|
- if (m#<PRE>#i) {
|
|
|
- s#^.*<PRE>#\\fC\n.DS#i;
|
|
|
- $preformat = 1;
|
|
|
- return 1;
|
|
|
- }
|
|
|
- s#<IMG SRC=.*>##i;
|
|
|
- if (m#<A NAME="toc">#i) {return 2};
|
|
|
- s#<A NAME=".*">##i;
|
|
|
- #s#<\!\-\-#.\\\" #i;
|
|
|
- #s#\-\->##i;
|
|
|
- if (m#^<HR.*>$#i) {return 0};
|
|
|
- s#<HR>##i;
|
|
|
- s#<HR .*>##i;
|
|
|
- if (&DoStrip($_)) {return 0}
|
|
|
- &DoFont($_);
|
|
|
- if (! &DoList($_)) {return 0};
|
|
|
- if (! &DoSection($_)) {return 0}
|
|
|
- &DoEscape($_);
|
|
|
- &DoPara($_);
|
|
|
- if (! $preformat) {
|
|
|
- if (m/^$/) {return 0};
|
|
|
- s#^[ \t]*##;
|
|
|
- s#<[^>]*>##g;
|
|
|
- }
|
|
|
- return 1;
|
|
|
-}
|
|
|
-
|
|
|
-sub DoEscape {
|
|
|
- s#>#>#i;
|
|
|
- s#<#<#i;
|
|
|
- s#&#&#i;
|
|
|
- s#"#"#i;
|
|
|
- s# # #i;
|
|
|
- s#©#©#i;
|
|
|
- s#ä#ä#i;
|
|
|
- s#ü#ü#i;
|
|
|
- s#<#<#i;
|
|
|
- s#>#>#i;
|
|
|
-}
|
|
|
-
|
|
|
-sub DoSection {
|
|
|
- if (m#^ *Table of Contents *$#i) {return 0};
|
|
|
- if (m#^<A NAME=".*">$#i) {return 0;}
|
|
|
- s#<A NAME=".*">##i;
|
|
|
- s#^.*<H2>#.SH #i;
|
|
|
- s#^.*<H3>#.SS #i;
|
|
|
- s#^.*<H4>#.SS #i;
|
|
|
- if (m#^</[^>]*>$#i) {return 0};
|
|
|
- s#</[^>]*>##ig;
|
|
|
- s#^<BR>$#.br#i;
|
|
|
- s#<BR>#\n.br\n#i;
|
|
|
- return 1;
|
|
|
-}
|
|
|
-
|
|
|
-
|
|
|
-#================
|
|
|
-# Paragraph breaks: <P>
|
|
|
-#
|
|
|
-sub DoPara {
|
|
|
- s/^ *<P>/.PP\n/i;
|
|
|
- s/<P>/\n.PP\n/ig;
|
|
|
- return;
|
|
|
-}
|
|
|
-
|
|
|
-#================
|
|
|
-# Lists
|
|
|
-# <DL><DT><DD>
|
|
|
-#
|
|
|
-sub DoList {
|
|
|
- if (m#^ *</DT> *$#i) {return 0};
|
|
|
- if (m#<DT> *$#i) {chop};
|
|
|
- s#^ *<DT>#.IP \"#i;
|
|
|
- s#<DT>#\n.IP \"#i;
|
|
|
- s#</DT>#\" 4m#i;
|
|
|
-
|
|
|
- if (m#^ *<DD> *$#i) {return 0};
|
|
|
- if (m#^ *</DD> *$#i) {return 0};
|
|
|
- if (m#^</LI>$#i) {return 0};
|
|
|
- s#<DD>#\n.br\n#i;
|
|
|
- s#</DD>##i;
|
|
|
-
|
|
|
- s#^.*<DL> *$##i;
|
|
|
- s#<DL>#\n.VL 4m\n#i;
|
|
|
-
|
|
|
- s#^.*</DL> *$#.PP#i;
|
|
|
- s#<DL>#\n.IP\n#i;
|
|
|
-
|
|
|
- s|<UL>|\n.RS\n|i and $in_bu=1;
|
|
|
- s|</UL>|\n.RE\n|i and $in_bu=0;
|
|
|
-
|
|
|
- s|<OL>|\n..IP\n|i and ($in_nb++,$nb_n[$in_nb]=1); # @nb_n is a stack allowing for nested <ol>'s
|
|
|
- s|</OL>|\n.PP\n|i and ($nb_n[$in_nb]=0,$in_nb--);
|
|
|
-
|
|
|
- s|</LI>||i;
|
|
|
- if(/<li>/){
|
|
|
- $in_bu and s|<LI>|\n.IP\n|i;
|
|
|
- $in_nb and ( # since we're hand-composing the ordered lists (not using mm's .AL),
|
|
|
- $n=(join ".", ( # we use a uniform indentation and x.y.z numbering
|
|
|
- (map {$_-1} @nb_n[1..($in_nb-1)]) # nb_n's are one beyond the current number for the levels above current
|
|
|
- ,
|
|
|
- $nb_n[$in_nb])), # and exact for the current level
|
|
|
- s|<li>|\n.IP \\fB$n\\fR\n|i,
|
|
|
- $nb_n[$in_nb]++ # because we post-increment after encountering each <li>
|
|
|
- );
|
|
|
- }
|
|
|
-
|
|
|
- # move into CvtTable() ??
|
|
|
- s/<TD.*?>/\t| /gi;
|
|
|
- s/<TR.*?>/\n.br\n/i;
|
|
|
-
|
|
|
- return 1;
|
|
|
-}
|
|
|
-
|
|
|
-#================
|
|
|
-# Font transitions:
|
|
|
-# <B> => \fB
|
|
|
-# </B> => \fR
|
|
|
-# <I> => \fI
|
|
|
-# </I> => \fR
|
|
|
-#
|
|
|
-sub DoFont {
|
|
|
- s# *</B> *_ *<B> *#_#ig;
|
|
|
- s#</B> *<B># #ig;
|
|
|
- s#<B>#\\fB#ig;
|
|
|
- s#</B>#\\fR#ig;
|
|
|
- s#<SUP>#\\u#ig;
|
|
|
- s#</SUP>#\\d#ig;
|
|
|
-
|
|
|
- s# *</I> *_ *<I> *#_#ig;
|
|
|
- s#</I> *<I># #ig;
|
|
|
- s#<I>#\\fI#ig;
|
|
|
- s#</I>#\\fR#ig;
|
|
|
- s# *</CODE> *_ *<CODE> *#_#ig;
|
|
|
- s#</CODE> *<CODE># #ig;
|
|
|
- s#<CODE>#\\fI#ig;
|
|
|
- s#</CODE>#\\fR#ig;
|
|
|
- s# *</EM> *_ *<EM> *#_#ig;
|
|
|
- s#</EM> *<EM># #ig;
|
|
|
- s#<EM>#\\fI#ig;
|
|
|
- s#</EM>#\\fR#ig;
|
|
|
- s/&NBSP;/ /ig;
|
|
|
- return;
|
|
|
-}
|
|
|
-
|
|
|
-#==================
|
|
|
-# Process HTML file
|
|
|
-
|
|
|
-$infile = $ARGV[0];
|
|
|
-$outfile = $ARGV[1];
|
|
|
-$preformat = 0;
|
|
|
-$page = &basename($ARGV[0], ".html");
|
|
|
-$sect = $ARGV[2];
|
|
|
-open(INP,$infile) || die "Can't open '$infile' for reading: $!";
|
|
|
-if (! &GetTitle($page,$sect)) {
|
|
|
- print "$infile: Did not find a valid <TITLE> line\n";
|
|
|
- close(INP);
|
|
|
- exit;
|
|
|
-}
|
|
|
-open(OUT,"> $outfile") || die "Can't open '$outfile' for writing: $!";
|
|
|
-print "Converting: $infile to $outfile\n";
|
|
|
-&PrintHeader($page,$sect,$TITLE[0]);
|
|
|
-
|
|
|
-while (<INP>) {
|
|
|
- # for intro pages which lack a module NAME (for whois entry)
|
|
|
- if ($_ =~ s/<!-- meta page description:(.*)-->/$1/) {
|
|
|
- print OUT ".SH NAME\n$page -$_\n";
|
|
|
- }
|
|
|
-
|
|
|
- s/\\/\\(rs/g;
|
|
|
- $result = &DoLine($_);
|
|
|
- if ($result == 2) {
|
|
|
- close(INP);
|
|
|
- close(OUT);
|
|
|
- return;
|
|
|
- }
|
|
|
-
|
|
|
- if (&DoLine($_) == 1) {
|
|
|
- s#<#<#g;
|
|
|
- s#>#>#g;
|
|
|
- s#^'#\\(cq#gm;
|
|
|
- s#^"#\\(dq#gm;
|
|
|
- s#`#\\(ga#g;
|
|
|
-
|
|
|
- #
|
|
|
- # this part removes the comments
|
|
|
- #
|
|
|
- $line = $_;
|
|
|
- $line =~ s/<!--.*-->//g; #
|
|
|
- if ($line =~ s/(.*)<!--.*/$1/) {
|
|
|
- $suppress = 1;
|
|
|
- $out .= $line;
|
|
|
- }
|
|
|
- if ($line =~ s/.*\-\-\>(.*)/$1/g) {
|
|
|
- print $line
|
|
|
- $suppress = 0;
|
|
|
- }
|
|
|
- if ($suppress != 0) {
|
|
|
- $line = "";
|
|
|
- }
|
|
|
- #
|
|
|
- # end of comment part
|
|
|
- #
|
|
|
-
|
|
|
- $out .= $line;
|
|
|
-
|
|
|
- if ($preformat == 1) {
|
|
|
- $out .= ".br\n";
|
|
|
- }
|
|
|
- }
|
|
|
-}
|
|
|
-$out =~ s/\n+/\n/gm;
|
|
|
-print OUT CvtTable($out);
|
|
|
-close(INP);
|
|
|
-close(OUT);
|
|
|
-
|
|
|
-exit;
|
|
|
-
|
Glynn Clements