#!/usr/bin/perl -w # # PLies Copyright 2001 Dobrica Pavlinusic # # this tool is based on SlideMaker and XLSies tool # split a all.htm into slide*.htm # # For more information please see presentation about this tool on # http://www.rot13.org/~dpavlin/presentations/XLSies_to_PLies/ # # Copyright 1999 World Wide Web Consortium, # (Massachusetts Institute of Technology, Institut # National de Recherche en Informatique et en # Automatique, Keio University). All Rights Reserved. # This program is distributed under the W3C's # Intellectual Property License. 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 W3C License # http://www.w3.org/Consortium/Legal/ for more details. # # ############################################################################## # # slidemaker credits: # # Stephan Montigaud - stephan@w3.org # created 970601 # modified by Pierre Fillault # check the documentation at http://www.w3.org/Talks/YYMMsub/ # # modified 19990505 Bert Bos: ALT text of prev/next arrows is now # "prev"/"next" rather than the title of the prev/next slide; looks better # in lynx. # # version: 4.14 - 19990719 # $Id: w3cburst.pl,v 1.41 1999/11/02 17:25:50 charles Exp $ # # # XLSies credits: # # Sami Lempinen - lempinen@iki.fi # http://www.snowman.sgic.fi/ssl/xslies/ # # Text::FastTemplate: # Robert Lehr - bozzio@the-lehrs.com use Text::FastTemplate 0.91; use strict; ############################################################################## ## default values of variables ## # save base dir my $base_dir = $0; $base_dir =~ s,/[^/]+$,,; # remove name of script ## show debug output my $debug=0; ## default DOCTYPE added on the slides my $doctype = ''; ## name of raw HTML file containing the slides my $all = 'all.htm'; ## table of content built from all.htm - also first page of the presentation ## this is only the basename as we need to generate one toc for each style sheets ## the main toc will not bear any more so the server can understand a request for '/' ## the next ones will bear a number corresponding to the slide index #$overview = 'Overview'; my $overview = 'index'; ## name of the file containing the parameters of the presentation my $infos = 'infos.txt'; ## link to the logo printed on all the slides my $logoLink = ''; ## default location of the logo - works when slidemaker is used as a package my $logoFile = ''; ### localization my $loc_toc = "Table of contents"; my $loc_by = "by"; my $loc_amp = "&"; # author separator my $loc_slide = "Slide"; my $loc_of = "of"; # $loc_slide nr $loc_of total ## alternate representation of the logo my $logoAlt = ''; ## default values set to none my $logoLink2 = ''; # link to a potential second reference my $logoFile2 = ''; # location of a second logo my $logoAlt2 = ''; # alternate representation of the second logo ## default accesskeys for navigation icons used in the slides my $prevKey = 'P'; # accesskey for previous slide my $nextKey = 'N'; # accesskey for next slide my $tocKey = 'C'; # accesskey for table of contents my $styleKey = 'S'; # accesskey for changing style sheets ## default author(s) name and URL my $author = 'Staff'; my $authorUrl; my $author2; my $author2Url; ## default presentation title my $talkTitle = 'Talk'; my $talkSubTitle; ## standard style sheets my $cssStandard = "$base_dir/css/default.css"; ## template name my $template = "$base_dir/default"; ## default charset use in meta tag http-equiv (undef to skip) #my $charset = 'ISO-8859-1'; my $charset; ## default tag my $body = ''; ## number of entries on each TOC page my $toc_on_page = 10; ## use progress bar my $progress_bar = 1; ## content hight for each slide my $content_hight = "70%"; ## some other vars which can be setuped in infos.txt my $date; ## end of default values for the presentation ############################################################################## ## globals my $last_toc_title = ""; my %page_data; my %overview_data; my $pack = 0; my @pack_additional; # additional files to pack (pictures, logos...) my %nesttag = init_nesttag(); my $http_equiv; my @toc_entries; # entries in TOC my $js_code = "$base_dir/pylize.js"; # add perl_include to @INC path push @INC, "$base_dir/perl_include" if (-e "$base_dir/perl_include"); ############################################################################## ## reading user input from $infos ## my @file; ############################################################################## sub parse_infos { my @file=@_; do { if ($file[0] && $file[0] =~ /^[^#\n\r]/) { $file[0] =~ s/\n//; # remove UNIX \n $file[0] =~ s/\r//; # remove WINDOWS \r my ($var,$value) = split(/ *= */,$file[0],2); $value=~s/'/\\'/g; my $cmd; if ($value ne "") { $cmd="\$$var = \'$value\';"; } else { $cmd="\$var = \'$value\' || 1;"; } if (defined($value)) { eval($cmd) || warn "problem with eval of: $cmd"; } else { die "no value defined for $var"; } print STDERR "$file[0]\n"; } } while (shift(@file)); } ############################################################################## parse_infos(@ARGV); # backward compatibility and for pack # read parameters from infos.txt and put them in @PARAM if (open(INFOS, $infos)) { print STDERR "--- Reading parameters file $infos ---\n"; @file = ; parse_infos(@file); } # try to read local infos.txt for template if (-f "$template/$infos" && open(INFOS,"$template/$infos")) { print STDERR "--- Reading template parameters file $template/$infos ---\n"; @file = ; parse_infos(@file); close(INFOS); } ## @PARAM is now a table with the user preferences for his presentation ## use charset if ($charset) { $http_equiv=''; } else { $http_equiv=''; } ############################################################################## ## read the raw html presentation ## ## copy file in memory my $sep = $/; $/ = undef; open(ALL, $all) || die "Error: Cannot open file: $all"; my $buf = ; $buf =~ s/\r//g; # remove WINDOWS \r close(ALL); $/ = $sep; ## Remove comments from the raw presentation ## they do not need to show up on the slides $buf =~ s///sgo; ## if $pack is set, output name of css (for inclusion in archive), but ## reset $cssStandard only to filename (without path) die "Error: Can't open css file: $cssStandard" if (!-r $cssStandard); die "Error: Can't open js file: $js_code" if (!-r $js_code); if ($pack) { push @pack_additional,$cssStandard; $cssStandard =~ s/^.*\/([^\/]+)$/$1/g; push @pack_additional,$js_code; $js_code =~ s/^.*\/([^\/]+)$/$1/g; } ## the slidemaker tool assumes that each slide is self contained between 2 sets of h1 tags ## if not it will generate a rather weird output ## split using and as separator (ignores attributes!) ## h1 or H1 can be used my @table = split(/<\/?[hH]1[^>]*>/, $buf); ## compute the total number of slides my $total = $#table / 2; if ($#table % 2 != 0) { $total = ($#table +1)/2; } ## ## raw presentation has been read successfully ############################################################################## ############################################################################## ## processing the slides print STDERR "\n--- Processing $total slides ---\n"; ## generate the header table of content of the presentation ## which is also the first page of the talk &openOverview($overview); ## start the slide count so we can number them my $slideCount = 1; ## pre-load template slides using $template dir Text::FastTemplate->defaults( path => [ $template ] ); Text::FastTemplate->preload( [ { file => 'slide.html', key => 'slide' }, { file => 'overview.html', key => 'overview' }, { file => 'title.html', key => 'title' } ] ); ## unroll relative anchors (#something) into links with slides my %anchor_on_slide; ## step 1: record anchors for(my $i=0; $i<=$total; $i++) { my $tmp = $table[($i*2)]; $tmp .= $table[($i*2)+1] if ($table[($i*2)+1]); while ($tmp =~ s,"]+)"*>,,i) { $anchor_on_slide{$1}=$i; print STDERR "\tslide ",($i+1)," anchor: $1\n" if ($debug); } } ## step 2: fix links for(my $i=0; $i<$total; $i++) { foreach (keys %anchor_on_slide) { $table[($i*2)] =~ s,href="*#$_"*>,href="slide$anchor_on_slide{$_}\.html#$_">,gi; } } ## @table is the array containing each slide with its title ## for each slide to be generated ## we delete each slide and its title when generated ## so that the current slide and its title are always at $table[0] (for the title) ## and $table[1] (for the slide content) #undef $prev_title; #undef $next_title; my $prev_title; my $next_title; do { ## get rid of the first element contained by the raw presentation array shift(@table); ## then $table[0] is the title of the slide to be generated if ($table[0]) { $table[0] =~ s/[\n\r\s]+/ /g; # compress whitespace $table[0] =~ s/^ +//; $table[0] =~ s/ +$//; } ## $slideTitle preserves link(s) in the title my $slideTitle = $table[0]; ## need to check if the title contains any anchor ## if so it needs to be removed ## because the title is being used in the table of content to link to the corresponding slide $table[0] = remove_anchor($table[0]); ## grab next slide title $table[2] (if there's a next slide) ## to be able to use in the 'next' navigation button ## keep in mind that $table[1] contains the slide corresponding to the title $table[0] if ($table[2]) { $next_title= remove_anchor($table[2]); } ## the current slide content is stored $table[1] ## there is an attempt to make sure it's clean HTML ## Pierre Fillault's note: use same piece of as used in http://www.w3.org/Web/Tools/CvsCommitScripting ## to make use of the validation service my $slideContent = &clean_html($table[1]) ; ## extract slide Sub Title

my $slideSubTitle; if ($slideContent && $slideContent =~ s/]>(.+?)<\/h2[^>]>//ism) { $slideSubTitle=remove_anchor($1); } ## add the title of the current slide to the table of content &addTitle($slideTitle,$slideSubTitle,$slideCount); ## generate the current slide ## parameters are: ## title of the slide, its content, the slide number, the title of the previous slide and the title of the next slide &createSlide($slideTitle,$slideSubTitle,$slideContent ,$slideCount++,$prev_title,$next_title); ## save the title of the previous slide to be displayed in the 'previous' navigation button $prev_title=remove_anchor($table[0]); } ## process the next slide while (shift(@table)); ## close the table of content &closeOverview; ## print additional files to pack print STDOUT join("\n",@pack_additional) if ($pack); print STDERR "--- Finished ---\n"; exit 0; ## ## end of the slidemaker main program ############################################################################## ############################################################################## ## generate the header of the table of content sub openOverview { my ($file) = @_; ## open the file to write to open(FOO, ">${file}.html") || die "can't open ${file}.html: $!"; push @pack_additional,"${file}.html" if ($pack); ## the style sheet used in the table of content is my $stylelink = ""; ## here is the standard style sheet $stylelink .= ""; # add javascript anim $stylelink .= head_additional(undef, "slide1.html"); %overview_data = ( doctype => $doctype, title => $talkTitle, http_equiv => $http_equiv, stylelink => $stylelink, body => $body, logoLink => $logoLink, logoFile => $logoFile, logoAlt => $logoAlt, logoLink2 => $logoLink2, logoFile2 => $logoFile2, logoAlt2 => $logoAlt2, talkTitle => $talkTitle, talkSubTitle => $talkSubTitle, content_hight => $content_hight, author => $author, authorUrl => $authorUrl, author2 => $author2, author2Url => $author2Url, date => $date, toc_title => $loc_toc, template_dir => "$template/", ); } ## ## the beginning of the table of content has been generated and saved ############################################################################## ############################################################################## ## generate the footer of the table of content sub closeOverview { $overview_data{slide_html} = make_progress_bar(0,$total); $overview_data{toc_entries} = [ @toc_entries ]; my $page= new Text::FastTemplate key => 'overview'; my $html = add_onLoad( $page->output( \%overview_data ) ); if ($pack) { print FOO extract_files($html, $overview_data{template_dir}); } else { print FOO $html; } close(FOO); } ## ## the toc has been completed and saved ############################################################################## ############################################################################## ## add an item in the toc sub addTitle { my $title = remove_tags(shift @_); my $subtitle = remove_tags(shift @_); my $nr = shift @_; if (! $title) { return 1; } # split TOC entries to multiple pages my $pre_ul; my $post_ul; if ($nr % $toc_on_page == 0) { my $toc_nr=int($nr/$toc_on_page); my $item = { pre_html => $pre_ul, accesskey => " ", # space href => "index-toc$toc_nr.html", title => "...", post_html => $post_ul, more => 1, # use style for more pages link (...) }; push @toc_entries, $item; &closeOverview; undef @toc_entries; &openOverview("${overview}-toc${toc_nr}"); $last_toc_title=''; } $pre_ul=$post_ul=''; if ($last_toc_title eq $title && $subtitle) { if ($subtitle !~ m,^\s$,) { $title = $subtitle; } else { $title .= " (...)"; } $pre_ul='
'; $post_ul='
'; } else { $last_toc_title=$title; } # add accesskey for first 9 slides (`1' - `9') or just for first # TOC page, and tabindex for all slides if ($nr < 10 && $nr < $toc_on_page) { my $item = { pre_html => $pre_ul, accesskey => "$nr", tabindex => "$nr", href => "slide$nr.html", title => $title, post_html => $post_ul, more => 0, }; push @toc_entries,$item; } else { my $item = { pre_html => $pre_ul, tabindex => "$nr", href => "slide$nr.html", title => $title, post_html => $post_ul, }; push @toc_entries,$item; } } ## ############################################################################## ############################################################################## ## generate the current slide sub createSlide { # parameters are respectively the slide title, its content, # its number, the next slide title and the previous slide title my $title = remove_tags(shift @_) || return 1; my $subtitle = remove_tags(shift @_); my $content = shift @_ || return 1; my $nr = shift @_; my $prev_title = remove_tags(shift @_); my $next_title = remove_tags(shift @_); # work the slide title, the previous and next titles chomp $title; # work the slide content $content =~ s/<\/body>//i; # remove if any $content =~ s/<\/html>//i; # remove if any print STDERR remove_tags(sprintf("Slide %2d: %s %s", $nr, $title, $subtitle || "")),"\n"; &verify_html($content); # check the html &check_tags($content); # check open and closed tags ## write to the slide open(SLIDE, ">slide$nr.html") || die "can't save slide$nr.html: $!"; push @pack_additional,"slide$nr.html" if ($pack); my $toc_link = "$overview\.html"; ## initialization of the navigation links my $next_link = ""; my $prev_link = ""; if ($nr>1) { $prev_link = "slide".($nr-1).".html"; # } else { # ## add a link back to the toc for the first slide --CMN 19991102 # $prev_link = "$overview\.html"; } if ($nr != $total) { $next_link = "slide".($nr+1).".html"; } my $stylelink = ""; # here is the standard style sheet $stylelink .= "\n"; $stylelink .= head_additional($prev_link, $next_link); # add javascript anim my $slide_html=make_progress_bar($nr,$total); # undefine body if no content is found (so that template can show # only title and sub-title if ($content !~ m/\S/g) { undef $content; } %page_data = ( doctype => $doctype, talkTitle => $talkTitle, title => $title, subtitle => $subtitle, http_equiv => $http_equiv, stylelink => $stylelink, body => $body, logoLink => $logoLink, logoFile => $logoFile, logoAlt => $logoAlt, logoLink2 => $logoLink2, logoFile2 => $logoFile2, logoAlt2 => $logoAlt2, content_hight => $content_hight, content => $content, prev_link => $prev_link, toc_link => $toc_link, next_link => $next_link, prev_title => $prev_title, toc_title => $loc_toc, next_title => $next_title, author => $author, authorUrl => $authorUrl, author2 => $author2, author2Url => $author2Url, date => $date, slide_html => $slide_html, template_dir => "$template/", ); my $page; if ($content) { $page= new Text::FastTemplate key => 'slide'; } else { $page= new Text::FastTemplate key => 'title'; } my $html = add_onLoad( $page->output( \%page_data ) ); if ($pack) { print SLIDE extract_files($html, $page_data{template_dir}); } else { print SLIDE $html; } close(SLIDE); return 0; } ############################################################################## # check that the html of the slide # is correct (ALT tags, ...) # This procedure produces only warning sub verify_html { if ($_[0] =~ /])>/im) { if (!($1 =~ /ALT=/im)) { print STDERR "WARNING: without ALT\n"; print STDERR " \n" ; } } } ############################################################################## # clean the html of the slide # remove all
blabla
sub clean_html { my $tmp=$_[0] || return; $tmp =~ s/]|\"comment\").?<\/div>//igs; $tmp =~ s,]>,,igms; # execute perl code inside ... tags # warning: this create one fork per tag! sub eval_perl { my $code = shift @_; my $pid = open(FH, "-|"); my $output; if ($pid) { # parent while () { $output .= $_; } close(FH); } else { # child eval($code); die " eval failed: $@ " if ($@); exit 0; } return $output; } $tmp =~ s,(.?),eval_perl($1),igmse; return $tmp; } ############################################################################## # make transparent 1x1 gif (for layout tricks) sub make_dotgif { my @dot_gif=(71,73,70,56,57,97,1,0,1,0,128,0,0,192,192,192,0,0,0,33, 249,4,1,0,0,0,0,44,0,0,0,0,1,0,1,0,0,2,2,68,1,0,59); open(GIF,"> $_[0]") || die "can't write gif $_[0]: $!"; my $dotgif; foreach (@dot_gif) { $dotgif.=chr($_); } print GIF $dotgif; close(GIF); } ############################################################################## # make slide progress bar sub make_progress_bar { my ($nr,$total) = @_; my $pcnt_done=int($nr100/$total); my $pcnt_left=100-$pcnt_done; my $html; if ($progress_bar && uc($progress_bar) ne "NO") { my ($l,$r); $l = $r = " "; my $t="$nr of $total"; if ($pcnt_done > 50) { $l=$t; } else { $r=$t; } $html=''; if ($pcnt_done != 0) { $html.=''; } if ($pcnt_left != 0) { $html.=''; } $html.='
'.$l.' '.$r.'
'; } else { $html="$loc_slide $nr $loc_of $total"; } return $html; } ############################################################################## # remove anchors from html (for titles) sub remove_anchor { my $tmp = $_[0] || return; $tmp =~ s/(.)]>(.)<\/A>(.)/$1$2$3/ig; return $tmp; } ############################################################################## # remove html tags sub remove_tags { my $tmp = shift @_ || return; $tmp =~ s/<[^>]+>//g; $tmp =~ s/\r//g; $tmp =~ s/"/"/g; # quote " return $tmp; } ############################################################################## # extract files referenced in presentation and remove dirs from slide html sub extract_files { my $slide = shift @_; my $template_dir = shift @_; my $tmp = $slide; while ($tmp =~ s/href="([^"\s]+)"//ism || $tmp =~ s/src="([^"\s]+)"//ism) { if ("$1" !~ m/[hf]t?tp:/ && -f "$1" && !grep(/$1/,@pack_additional)) { my $file=$1; push @pack_additional,$file; if ($file =~ m,^(.+)/([^/]+),) { my ($d,$f) = ($1,$2); $slide =~ s,${d}/${f},${f},g; } } } $slide =~ s,$template_dir,,smg; return $slide; } ############################################################################## # check tags in slide # based on code from hindent 1.1.2 by Paul Balyoz sub init_nesttag { # Tags that require their own end tag ... we will nest them # properly: (WARNING, you must use lower-case here) # All other tags (not on this list) will be ignored for indenting purposes. return ( 'html' => 1, 'head' => 1, 'body' => 1, 'title' => 1, 'a' => 1, 'table' => 1, 'tr' => 1, 'th' => 1, 'td' => 1, 'form' => 1, 'select' => 1, 'textarea' => 1, # 'p' => 1, Don't do this one because many people use
but not
'ul' => 1, 'ol' => 1, 'dl' => 1, 'blockquote' => 1, 'center' => 1, 'div' => 1, 'font' => 1, 'pre' => 1, 'tt' => 1, 'i' => 1, 'b' => 1, 'u' => 1, 'strike' => 1, 'big' => 1, 'small' => 1, 'sub' => 1, 'sup' => 1, 'em' => 1, 'strong' => 1, 'dfn' => 1, 'code' => 1, 'samp' => 1, 'kbd' => 1, 'var' => 1, 'cite' => 1, 'h1' => 1, 'h2' => 1, 'h3' => 1, 'h4' => 1, 'h5' => 1, 'h6' => 1, 'applet' => 1, 'map' => 1, 'frameset' => 1, 'noframes' => 1, ); } sub check_tags { my $tmp = $_[0]; my @tagstack; my $level=0; while ($tmp =~ /<([^>]?)>/gms) { my $tag=$1; $tag=~s/\s.$//g; # if regular tag, push it on stack; if end-tag, pop it off stack. # but don't do any of this if it's not a special "nesting" tag! if ($tag !~ m,^/,) { if ($nesttag{lc($tag)}) { push @tagstack,$tag; $level++; # remember how much for later } } else { $tag =~ s,^/,,; # convert this end-tag to a begin-tag $tag = lc($tag); if ($nesttag{lc($tag)}) { # throw away tags until we find a match if ($#tagstack > -1) { while ($tag ne lc(pop @tagstack)) { $level--; # we threw away extra tags last if $#tagstack <= 0; } $level--; # we threw away extra tags if ($level < 0) { print STDERR "WARNING: more end than begin tags around !\n"; } } } } } if ($level > 0) { print STDERR "WARNING: level=$level, ", $#tagstack+1," tags left on stack after done parsing! Specifically:\n<",join("> <",@tagstack),">\n"; } } ############################################################################## # # add javascript support from pylize by # Christopher Arndt - chris.arndt@web.de # http://www.chrisarndt.de/en/software/pylize/ sub head_additional { my $prev_link = shift @_; my $next_link = shift @_; my $html; $html .= "\n"; $html .= "\n" if ($prev_link); $html .= "\n" if ($next_link); # add javascript support from pylize by # Christopher Arndt - chris.arndt@web.de # http://www.chrisarndt.de/en/software/pylize/ $html .= ' '; return $html; } sub add_onLoad { my $html = shift @_; $html =~ s/(