#!/usr/bin/perl -w use strict; #-------------------------------------------------------- # sbib2html.pl # # author: saar drimer, .@cl.cam.ac.uk # # webpage: http://www.saardrimer.com # http://www.cl.cam.ac.uk/~sd410 # # project: http://www.cl.cam.ac.uk/~sd410/fpgasec # # See my comments in the code for guidance. In brief, if # you want to order your entries you need to add a # "www_section = {foo}," field to each entry. The # ordering of the _categories_ is determined by the # first instance in which they appear, then by sequential # order in which they are ordered in the bibtex file. If # you run into trouble consult the raw bibtex file linked # in the project's webpage. In a case of emergency, # call 911 (or your local equivalent), then contact me. # # Copyright notice: you are welcome to use this script to # your heart's content... but please do give some credit # and link back. Be nice. It is supplied as-is. I believe # in the creativity of people; therefore, someone might # find a way to hurt themselves using this code. In that # case I am wholly void of responsibility. Don't forget # your safety goggles... a pint to the first person who # emails me the words "Stanley Spadowski"! # # I would also appreciate any comments, suggestions and # corrections. # #-------------------------------------------------------- print < End exit 1 if $#ARGV < 0; $ARGV[0] =~ s/\..*//; my $infile = "$ARGV[0].bib"; # input file my $outfile = "$ARGV[0].html"; # output html file my $outbib = "$ARGV[0]_web.bib"; # output bib file for web consumption # slurp file local( *IN ) ; open( IN, $infile ) or die "Can not open $infile\n"; my $fc = do { local( $/ ) ; }; close (IN); # remove all comments that start in a new line; this will # avoid removing '%' in URLs for example $fc =~ s/^%.*\n//g; # separate all entries into array my @bibs = $fc =~ /(\@.+?\})\s*\n/smg; open (OUT, ">$outfile"); open (WEB, ">$outbib"); my $i = 0; # this is a dump of an external html header information and # tags. use it to provide the mandatory html tags, meta data # css/script calls etc... up to and including the tag. open (TMP, "head"); while () { print OUT; } close TMP; # print header for web version of bib file print WEB "%------------------------------------------\n% FPGA security bibliography\n% created by: saar drimer\n% http://www.cl.cam.ac.uk/users/sd410\n% http://www.saardrimer.com\n%------------------------------------------\n\n"; my @cat = (); my %cat = (); my $tmp; my $total= 0; # get section names foreach my $b (@bibs) { ($tmp) = ($b =~ /www_section\s*=\s*\{([^,\}]*)[,\}]/); # all entried must have a "www_section" categoty. if (!defined($tmp)) { die "\nERROR: no \'www_section\' category found for the following entry:\n$b\n";; } else { # if "www_section" = 'none' then the entry will not be processed if ($tmp ne 'none') { $total++; push @{$cat{$tmp}}, $b; # this is needed to get the categories ordered by when they first # appear in the bib file. You can not determine the ordering # of a hash (and sorting isn't quite what I wanted) if (!(grep (/$tmp/, @cat))) { push (@cat, $tmp); } } } } $i = 1; my $hi; my $type; my $author; my $title; my $org; my $year; my $section; my $url; my $abstract; my $recs; print OUT "\n

categories | "; foreach my $k (@cat) { printf OUT "%s | ", $k, $k; } print OUT "

\n"; print OUT "

contract all | expand all

legend: FPGA specific, general security (click $\"expand\"$ for abstract and bibtex entry) [$total entries]

"; foreach my $k (@cat) { # print categoty header for web version of bib file printf WEB "%%------------------------------------------\n%% category: %s\n%%------------------------------------------\n\n", $k; $recs = @{$cat{$k}}-1; printf OUT "\n

%s

(contract all | expand all)

\n", $k, $k, $i, $i+$recs, $i, $i+$recs; foreach my $b (@{$cat{$k}}) { printf WEB "%s\n\n", $b; # HTML validators complain when they see '&' in the URL. Therefore, # it needs to be changed to '&' ($b) =~ s/\&/\&\;/g; ($type) = ($b =~ /\@(.*)\{/); ($author) = ($b =~ /author\s*=\s*\{(.*?)\},\s*\n/); ($title) = ($b =~ /title\s*=\s*\{(.*?)\},\s*\n/); ($org) = ($b =~ /organization\s*=\s*\{(.*?)\},\s*\n/); ($year) = ($b =~ /year\s*=\s*\{(.*?)\},\s*\n/); ($section) = ($b =~ /www_section\s*=\s*\{(.*?)\},\s*\n/); ($url) = ($b =~ /url\s*=\s*\{(.*?)\},\s*\n/); ($abstract) = ($b =~ /abstract\s*=\s*\{(.*?)\},\s*\n/s); if ($author) { ($author) =~ s/\{\\.\{(.)\}\}/$1/g; # replace accents ($author) =~ s/\{\\ss\{\}\}/ss/g; # replace \ss{} accent ($author) =~ s/[\{\}]//g; ($author) =~ s/ and /, /g; ($author) =~ s/(.*),/$1 and/; } # some entries, like web pages, do not have a title (the reason that what # would be the title appear in the author field is because it makes it easier # for sorting by the .bst definition file); this uses the author as the # title for placing the link. if ($title eq '') { $title = $author; } ($title) =~ s/\\EUR\{(.*?)\}/&\#8364 $1/g; ($title) =~ s/[\{\}]//g; if ($abstract) { ($abstract) =~ s/\n/
/g; } # remove abstract from bib entry $b =~ s/ *abstract\s*=\s*\{.*\}, *\n//smg; $b =~ s/\n/\/g; # highlight item if it is FPGA specific if ($section =~ /FPGA/) { $hi = ' style="background-color: #fffacd;"'; } else { $hi = ''; } printf OUT "

\n $\"contract\"$ \n", $hi; if ($author) { printf OUT "%s: ", $author;} elsif ($org) { printf OUT "%s: ", $org;} if ($url) { printf OUT "\"%s\"", $url, $title;} else { printf OUT "\"%s\"", $title;} if ($year) { printf OUT ", %s.", $year}; printf OUT "\n

\n"; # output abstract if exist if ($abstract) { printf OUT "", $abstract; } # output bib entry for cut-and-paste printf OUT "

Abstract. \n%s\n

", $b; $i++; } } print "processed $total entries\n"; # provide closing tags and info open (TMP, "foot"); while () { print OUT; } close TMP; my ($y, $m, $d) = (localtime)[5,4,3]; $y += 1900; $m += 1; print OUT "

last modified $y\/$m\/$d
\n"; print OUT ''; close (OUT); close (WEB);