$doc_title

#!/usr/bin/perl use strict; use Getopt::Std; use Encode qw/from_to/; my %opts; $opts{l} = 2; getopts('f:t:cpl:',\%opts); my @tree; my $last_title; my $last_content; my $last_level; my @whole_content = <>; my %LOCALVAR = find_localvar(\@whole_content); $opts{p} = 1 if $LOCALVAR{"document-multipage-mode"}; my $DEFAULT_CODING_SYSTEM = "big5"; $LOCALVAR{"buffer-file-coding-system"} ||= $DEFAULT_CODING_SYSTEM; $LOCALVAR{"document-output-coding-system"} ||= $LOCALVAR{"buffer-file-coding-system"}; unless ($LOCALVAR{"document-output-coding-system"} eq $LOCALVAR{"buffer-file-coding-system"}) { foreach (@whole_content) { from_to($_, $LOCALVAR{"buffer-file-coding-system"}, $LOCALVAR{"document-output-coding-system"}); } foreach (keys %LOCALVAR) { from_to($LOCALVAR{"$_"}, $LOCALVAR{"buffer-file-coding-system"}, $LOCALVAR{"document-output-coding-system"}); } } ## Expend [include:blah.txt], non-recursively ## $/ = undef; foreach (@whole_content) { if(m/^\s*\[include:(.*)\]\s*$/) { open(INCF, "< $1") || die ("Failed to insert the content of \"$1\""); $_ = ; close(INCF); } } @whole_content=map{"$_\n"}split/\n/,join"",@whole_content; ############################### foreach(@whole_content) { if (m/^(\*+)\s*(.+)$/) { push @tree, { title => $last_title, content => $last_content, level => $last_level }; $last_level = length($1); $last_title = $2; $last_content = ''; } else { $last_content .= $_; } } push @tree, { title => $last_title, content => $last_content, level => $last_level }; shift @tree; # the first one is empty. my $doc_title = $opts{t} || $LOCALVAR{"document-title"} || $ARGV[0]; my $htmlcharset = $LOCALVAR{"document-output-coding-system"}; my $htmlhead =qq{ $doc_title }; # Generate Index. my ($title,$before,$after,$close_last,$toc_string); $last_level = 1; my $root_i = -1; $toc_string = "

$_->{title}

"; $after = "

"; } elsif ($_->{level} < $last_level) { $close_last = "

"; } my $this_title = "${close_last}${before}${title}${after}"; $toc_string .= "${this_title}\n"; $last_level = $_->{level}; } $toc_string .= "

\n\n"; # Generate a div foreach section my @sections_content; my $section_content; $section_content = "

\n"; for my $i (0..$#tree) { $_ = @tree[$i]; if ($_->{level} == 1) { if ($i > 0) { $section_content .= "

\n"; $section_content .= " \n" if $opts{p}; push @sections_content,$section_content; $section_content = ''; $section_content = "

\n" if $opts{p} } $section_content .= "

\n"; } else { $section_content .= "

\n"; } $section_content .= "{level}>$_->{title}{level}>"; $section_content .= gencontent($_->{content}) . "\n" if ($_->{content} =~ /[^\s]/); $section_content .= "

\n" unless($_->{level} == 1); } $section_content .= "

\n"; $section_content .= "

\n"; push @sections_content,$section_content; my $footer = $opts{f} || $LOCALVAR{"document-footer"} || 'This page is generated by outline2html'; my $htmlbutt; if ($footer) { $htmlbutt = qq{ \n}; } else { $htmlbutt = qq{\n}; } if ($opts{p}) { for my $i (0..$#sections_content) { $_ = $sections_content[$i]; open(FH,">root-$i.html"); print FH $htmlhead; if ($opts{c}) { print FH "\n"; } else { print FH "\n"; } print FH $_; print FH $toc_string; print FH $htmlbutt; close(FH); } } else { print $htmlhead; print $toc_string; print foreach(@sections_content); print $htmlbutt; } # Other subroutines sub gencontent { my $content = shift; my @paragraphs = split(/\n\n\n*/,$content); # enum, list

foreach (@paragraphs) { s/^\n+//s; s/\s+\n//s; # Escape html entitites, but not those in the [ ]. # http://www.gugod.org/ => http://www.gugod.org my $urlpattern = '(?:http|https|ftp)://(?:[^/\[\]:]+)(?::\d+)?/[^\|\s\[\]\<\>]*'; my $emailpattern= '(?:[\w\d\.-]+@[\w\d\.-]+)'; use English; my @lines = split /\n/; my $prefixed = 0; foreach(@lines) { if(/$urlpattern/) { my($pre,$post) = ($PREMATCH, $POSTMATCH); unless(($pre =~ /\[/ && $post =~ /\]/) || ($pre =~ /$/ && $post =~ /$/)) { s{($urlpattern)}{[\1|\1]}; } } elsif (/$emailpattern/ ) { my($pre,$post) = ($PREMATCH, $POSTMATCH); unless($pre =~ /\[/ && $post =~ /\]/) { s{($emailpattern)}{[link:\1,mailto:\1]}; } } } $_ = join("\n", @lines); # It could be considered as everyhing in a [ ] are left untouched. s{(?([^\[\]]*?)(?!\])}{\1>\2}g; # Handle special instructions s{\[image:\s*(.+?)\s*\]}{}ig; s{\[link:\s*(\S+?)\s*,\s*(.+?)\s*\]}{\1}ig; s{\[(${urlpattern})\|(.+?)\]}{\2}ig; s{([^\s]+?)\s*$(${urlpattern})$}{\1}ig; # Else, Treat anything in [ ] untouched. s{\[(.+?)\]}{\1}g; @lines = split /\n/; my $prefixed = 0; foreach(@lines) { $prefixed++ if(/^\s*[\-\.ox#@\+=]\s/); } if(($prefixed-1 == $#lines) && $#lines > 0) { my $new_p; foreach (@lines) { s{^\s*[\-\.ox#@\+=]\s(.+)$}{
\1

\n${new_p}

] ]> foo.txt > foo.html outline2html.pl S<[-p]> S<[B<-l> [I] ]> foo.txt =head1 OPTIONS -c Collapse indexes. Expend it only when necessary. -l Levels to expand in Table of content. Default to 2 -p Multi-page mode. (local var: document-multipage-mode) -t Title of this document. (local var: document-title) -f A constant footer on each page (local var: document-footer) =head1 DESCRIPTION This script is quite convienent for a Emacs outline-mode user who wants to make html pages quickly. It generates html pages in two styles: all nodes in a single page or a page for each root nodes. HTML pages are generate in a structured fashion with default CSS stylesheet named B. It is also possible to embed links and images , using the following format in your outline-mode text file: [image: picure.jpg] [link: link_title , path/to/some/where/] [http://url.of.some/where/|link_title] ... link_title (http://url.of.some/where/) ... The first one makes a

tag, where both of the second and the third make a link_title tag. Notice that ONLY the [link: ...] style is capable to accpet relative url. You may also simply write an URL or email address _in one line_ , it will automatically create a hyperlink for it. For example: http://foo.org/ => http://foo.org/ => >bar@no.com> PLEASE NOTICE YOU HAVE TO PUT A TRAILING / IF YOUR URL IS NOT ENDED WITH SOME FILENAME. http://foo.org won't work. Paragraphs are seperated by a blank line in outline text file, and will be convert to a

...

in html file. Enumerations could be done by this simple way: - the item 1 - the item 2 - the item 3 , this should be enough for an example You don't have to use the prefix '-' , just put the same string before each line. . the item 1 . the item 2 . the item 3 , this should be enough for an example A paragraph contain only lines like this will be convert to a

foobar

] special command to do so. The expended document would be as if the content of is inserted directly. However the last LocalVar section cannot be put into a seperate file and then include. Also notice that this includtion is non-recursive. The latest relase information would be available on =head1 KNOWN BUGS If you have a unordered list at the final paragraph of your text, the last item goes into a new paragraph. The [ ] verbatim scheme may be incomplete, randomly whitespaces may eats your hyperlink. And it CANNOT CROSS LINES. A pair of [ ] for verbatim purpose in different lines has no effect. And therefore may cause unexpected output. =head1 Copyright Copyright 2003 by Kang-min Liu . This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself. See

$doc_title

Table of Contents

foobar

foobar

foobar

foobar