Bundle-PBib
view release on metacpan or search on metacpan
lib/Biblio/bp/lib/bp-cs-html.pl view on Meta::CPAN
#
# bibliography package for Perl
#
# HTML character set.
#
# This uses the HTML 2.0 spec, as described in
# http://www.w3.org/hypertext/WWW/MarkUp/html-spec/html-spec_toc.html
# specifically,
# http://www.w3.org/hypertext/WWW/MarkUp/html-spec/html-spec_5.html#SEC5.7
# and
# http://www.w3.org/hypertext/WWW/MarkUp/html-spec/html-spec_9.html#SEC9.7
#
# At some point I will either make an html3 charset or add it as an option
# to this one. References for those characters may be found in
# http://www.w3.org/hypertext/WWW/MarkUp/html3/mathsym.html
# Also see
# http://www.to.icl.fi/~aj/html3_test/entities.html
#
# Big additions in HTML3 include the greek characters and proper typesetting
# spaces.
#
# As referenced in:
# http://www.acl.lanl.gov/HTML_WG/html-wg-96q1.messages/0102.html
# I am going to allow the &U+xxxx; form for unicode characters for HTML 3.
# I'm not sure if this is going to stay in the standard however.
#
# Specifically, I use the named references as opposed to the numeric, as that
# is recommended in the html3 draft spec, and it is much easier to read.
# Reading through the working notes of the HTML_WG, it looks like they
# specifically recommend translating Numerical Character Entities of Latin-1
# into the real things. See reference:
# http://www.acl.lanl.gov/HTML_WG/html-wg-95q1.messages/0920.html
# about the can of worms this opens. Specifically, &215; refers to _different_
# characters depending on which ISO 8859-x group you're working with! The
# worst part of it is that netscape understands all the NCEs but skips most
# of their names!
#
# For this last reason (many browsers don't properly handle named entities),
# I've decided to stay with ISO-8859-1 8bit characters.
#
# This really ought to be called cs-html-8859-1 or something similar, since
# we're doing all our characters as ISO-8859-1. What we really ought to
# do is seperate all the named entities and meta characters out from the
# actual encoded characters, then have a setting for the particular
# encoding. One really ought to look at:
# http://www.ics.uci.edu/pub/ietf/html/draft-ietf-html-i18n-03.txt
# But I'm not going to worry about it at the moment.
#
# XXXXX Todo: handle previous fonts in a more sane fashion.
#
# Dana Jacobsen (dana@acm.org)
# 5 July 1995 (last modified 17 March 1996)
package bp_cs_html;
######
$bib'charsets{'html', 'i_name'} = 'html';
$bib'charsets{'html', 'tocanon'} = "bp_cs_html'tocanon";
$bib'charsets{'html', 'fromcanon'} = "bp_cs_html'fromcanon";
# This is a regex to search for. If it succeeds, then we call the routine.
# otherwise not. If it is set to undef, the routine is always called.
$bib'charsets{'html', 'toesc'} = "[<\&]";
$bib'charsets{'html', 'fromesc'} = "[\&<>]|${bib'cs_ext}|${bib'cs_meta}";
######
#
# If you want HTML 3 characters
#
$opt_html3 = 0;
# Extra things like sub and sup.
$opt_htmlplus = 1;
$cs_init = 0;
$cs_to_init = 0;
$cs_fr_init = 0;
######
lib/Biblio/bp/lib/bp-cs-html.pl view on Meta::CPAN
# HTML 2.0 markup and the bp-meta equivalent
%metamap_2 = (
'0102', 'I',
'0103', 'B',
'0104', 'TT',
'0112', '/I',
'0113', '/B',
'0114', '/TT',
'1100', 'P',
'1101', 'PRE',
'1102', 'ADDRESS',
'1103', 'BLOCKQUOTE',
'1110', '/P',
'1111', '/PRE',
'1112', '/ADDRESS',
'1113', '/BLOCKQUOTE',
'1300', 'LI',
'1301', 'UL',
'1311', '/UL',
'1302', 'OL',
'1312', '/OL',
'2100', 'CITE',
'2101', 'EM',
'2102', 'VAR',
'2103', 'STRONG',
'2104', 'CODE',
'2105', 'KBD',
'2106', 'SAMP',
'2110', '/CITE',
'2111', '/EM',
'2112', '/VAR',
'2113', '/STRONG',
'2114', '/CODE',
'2115', '/KBD',
'2116', '/SAMP',
);
# HTML+ markup and the bp-meta equivalent
%metamap_p = (
'0302', 'SUP',
'0312', 'SUB',
'030F', '/SUP',
'031F', '/SUB',
);
# HTML 3.0 markup and the bp-meta equivalent
%metamap_3 = (
'0200', 'BIG',
'020F', '/BIG',
'0210', 'SMALL',
'021F', '/SMALL',
);
$cs_init = 1;
}
sub init_cs_fr {
&init_cs unless $cs_init;
# XXXXX We should just use unicode_approx.
# Map various unicode entities to our stuff.
%charmap_from = (
'2212', '-',
'2013', '--',
'2014', '---',
'2002', ' ', # These two are probably wrong.
'2003', ' ',
);
# HTML 2.0 Secondary meta mappings for fromcanon
# Note that these do _not_ get wrapped in <> like metamap_2 does.
%metafrs_2 = (
'2200', '<A ',
'2210', '</A>',
'2300', 'href="',
'2310', '">',
);
$cs_fr_init = 1;
}
sub init_cs_to {
local($f);
&init_cs unless $cs_init;
# HTML 2.0 entity names that map to ISO-8859-1 codes.
# This is used in tocanon only. We leave the rest in 8 bits because it
# seems every browser can display that, but many of them don't map the names.
%entitynames = (
'nbsp', 160,
'iexcl', 161,
'cent', 162,
'pound', 163,
'curren', 164,
'yen', 165,
'brvbar', 166,
'sect', 167,
'uml', 168,
'copy', 169,
'ordf', 170,
'laquo', 171,
'not', 172,
'shy', 173,
'reg', 174,
'hibar', 175,
'deg', 176,
'plusmn', 177,
'sup2', 178,
'sup3', 179,
'acute', 180,
'micro', 181,
'para', 182,
'middot', 183,
'cedil', 184,
'sup1', 185,
'ordm', 186,
'raquo', 187,
'frac14', 188,
'frac12', 189,
lib/Biblio/bp/lib/bp-cs-html.pl view on Meta::CPAN
'Uuml', 220,
'Yacute', 221,
'THORN', 222,
'szlig', 223,
'agrave', 224,
'aacute', 225,
'acirc', 226,
'atilde', 227,
'auml', 228,
'aring', 229,
'aelig', 230,
'ccedil', 231,
'egrave', 232,
'eacute', 233,
'ecirc', 234,
'euml', 235,
'igrave', 236,
'iacute', 237,
'icirc', 238,
'iuml', 239,
'eth', 240,
'ntilde', 241,
'ograve', 242,
'oacute', 243,
'ocirc', 244,
'otilde', 245,
'ouml', 246,
'divide', 247,
'oslash', 248,
'ugrave', 249,
'uacute', 250,
'ucirc', 251,
'uuml', 252,
'yacute', 253,
'thorn', 254,
'yuml', 255,
);
# Construct the backwards map for these unique characters
foreach $f (keys %metamap_2) {
$bmetamap_2{$metamap_2{$f}} = $f;
}
# HTML 2.0 Secondary meta mappings for tocanon
%metatos_2 = (
);
$cs_to_init = 1;
}
#####################
# Essentially we're ISO-8859-1, but with a few protected characters
# and some extra ones.
# XXX protect lonely & characters
sub tocanon {
local($_, $protect) = @_;
local($repl, $can, $mine, $unicode);
&bib'panic("cs-html tocanon called with no arguments!") unless defined $_;
# Check to see if we have anything to do.
return $_ unless /<|\&/;
# XXXXX Ignore named Latin-1 for now.
# Handle links first.
local($link, $text);
while (/<A\s+href="([^"]+)">(.*)<\/A>/i) {
($link, $text) = ($1, $2);
# Why, oh why, couldn't perl4 have had minimal matching? I HATE this.
$text =~ s/<\/A>.*//i;
s/<A\s+href="$link">$text<\/A>/${bib'cs_meta}2200${bib'cs_meta}2300$link${bib'cs_meta}2310$text${bib'cs_meta}2210/gi;
}
# XXXXX Zap these for now
s/<A[^>]*>//gi;
s/<BASE[^>]*>//gi;
s/<\/A>//gi;
s/<\/?(BODY|HEAD|HTML|TITLE)>//gi;
s/<\/?H\d[^>]*>//gi;
s/<IMG[^>]*>//gi;
s/<(BR|HR)>//gi;
return $_ unless /<|\&/;
&init_cs_to unless $cs_to_init;
# XXXXX We've got two ways we can approach this:
#
# 1) we can assume that special characters are rare, so we just
# walk through our maps in some sort of liklihood order, trying
# to replace things. We should run lots of tests only rarely.
# 2) we can assume that special characters are likely, so we want
# to loop over the characters in the string. That means we have
# to have both a forward and a backward map. More memory and more
# startup time are traded for increased performance.
#
# Which approach should we use? Maybe it depends on the charset?
#
# I'm taking a hybrid approach. Method 1 for HTML3 and HTML+. M2
# for meta strings (<...>). M2 for standard escapes, and it gets
# more complicated from there.
if ($opt_html3) {
while (($can, $mine) = each %metamap_3) {
$mine =~ s/(\W)/\\$1/g;
s/<$mine>/${bib'cs_meta}$can/g;
}
}
if ($opt_htmlplus) {
while (($can, $mine) = each %metamap_p) {
$mine =~ s/(\W)/\\$1/g;
s/<$mine>/${bib'cs_meta}$can/g;
}
}
while (/<(\S[^>]*)>/) {
$repl = $1;
# added uc() to support lower-case HTML tags (ptandler, 02-03-24)
defined $bmetamap_2{uc($repl)} && s/<$repl>/${bib'cs_meta}$bmetamap_2{uc($repl)}/g
&& next;
&bib'gotwarn("Unknown HTML markup: <$repl>");
s/<$repl>//g;
}
$_ = &bib'font_check($_) if /${bib'cs_meta}01/o;
# Convert html escapes for ISO-8859-1 things to standard form.
s/</</g;
s/>/>/g;
s/"/"/g;
# To prevent matching this, we leave it in extended format.
s/&/${bib'cs_ext}0026/go;
# always check to see if we have any characters left to change
return $_ unless /\&/;
# This should handle 8 bit chars correctly as well as the new UCS-2, cf.
# <http://www.ics.uci.edu/pub/ietf/html/draft-ietf-html-i18n-03.txt>
while (/\&#(\d+);/) {
$repl = $1;
$can = &bib'unicode_to_canon( &bib'decimal_to_unicode($repl) );
s/\&#$repl;/$can/g;
}
# This handles to the new &U+xxx; form. I need to find a good reference
# for this.
while (/\&U\+([\dA-Fa-f]{4});/) {
$repl = $1;
$can = &bib'unicode_to_canon( $repl );
s/\&U\+$repl;/$can/g;
}
# Now for the named entities.
while (/\&(\S*);/) {
$repl = $1;
$repl =~ s/(\W)/\\$1/g;
if (defined $entitynames{$repl}) {
$can = pack("C", $entitynames{$repl});
s/\&$repl;/$can/g;
next;
}
# We now search through the charmaps to see if they contain our character.
# Terribly inefficient, but we shouldn't be doing this often.
if (defined $opt_html3) {
$can = undef;
foreach $unicode (keys %charmap_3) {
next unless $charmap_3{$unicode} eq $repl;
# only here if we found the map.
$can = &bib'unicode_to_canon($unicode);
s/\&$repl;/$can/g;
last;
}
next if defined $can;
}
$can = undef;
foreach $unicode (keys %charmap_2) {
next unless $charmap_2{$unicode} eq $repl;
# only here if we found the map.
$can = &bib'unicode_to_canon($unicode);
s/\&$repl;/$can/g;
last;
}
next if defined $can;
&bib'gotwarn("Unknown HTML entity: \&$repl; : in $_");
s/\&$repl;//g;
}
s/&(\s)/${bib'cs_ext}0026$1/g;
return $_ unless /\&/;
&bib'gotwarn("Unknown HTML characters in '$_'");
$_;
}
######
sub fromcanon {
local($_, $protect) = @_;
local($repl, $can);
&bib'panic("cs-html fromcanon called with no arguments!") unless defined $_;
# Leave 8bit characters alone since we're assuming ISO-8859-1 HTML.
#1 while s/${bib'cs_ext}(00..)/&bib'unicode_to_canon($1)/ge;
# XXXXX I think these should go here, as they create a mess if they go
# after the ext and meta maps.
s/\&/\&/g;
s/</\</g;
s/>/\>/g;
s/${bib'cs_ext}0026/\&/go;
return $_ unless /$bib'cs_escape/o;
&init_cs_fr unless $cs_fr_init;
# Extension characters
while (/${bib'cs_ext}(....)/o) {
$repl = $1;
if ($repl =~ /^00/) { # The 8bit mapping is the same
1 while s/${bib'cs_ext}00(..)/pack("C", hex($1))/goe;
next;
}
$opt_html3 && defined $charmap_3{$repl}
&& s/$bib'cs_ext$repl/\&$charmap_3{$repl};/g
&& next;
defined $charmap_2{$repl}
&& s/$bib'cs_ext$repl/\&$charmap_2{$repl};/g
&& next;
defined $charmap_from{$repl}
&& s/$bib'cs_ext$repl/$charmap_from{$repl}/g
&& next;
$opt_html3 && s/$bib'cs_ext$repl/\&U\+$repl;/g && next;
$can = &bib'unicode_approx($repl);
defined $can && s/$bib'cs_ext$repl/$can/g && next;
&bib'gotwarn("Can't convert ".&bib'unicode_name($repl)." to HTML");
s/${bib'cs_ext}$repl//g;
}
# XXXXX We need to deal with font changes.
$_ = &bib'font_noprev($_) if /${bib'cs_meta}0110/o;
while (/${bib'cs_meta}(....)/o) {
$repl = $1;
# We need to do these in order of most to least inclusive.
# That way we can get, say ThinSpace to map to a thin space in HTML3
# and a space in HTML2.
$opt_html3 && defined $metamap_3{$repl}
&& s/${bib'cs_meta}$repl/<$metamap_3{$repl}>/g
&& next;
$opt_htmlplus && defined $metamap_p{$repl}
&& s/${bib'cs_meta}$repl/<$metamap_p{$repl}>/g
&& next;
defined $metamap_2{$repl}
&& s/${bib'cs_meta}$repl/<$metamap_2{$repl}>/g
&& next;
defined $metafrs_2{$repl}
&& s/${bib'cs_meta}$repl/$metafrs_2{$repl}/g
&& next;
$can = &bib'meta_approx($repl);
defined $can && s/$bib'cs_meta$repl/$can/g && next;
&bib'gotwarn("Can't convert ".&bib'meta_name($repl)." to HTML");
s/${bib'cs_meta}$repl//g;
}
$_;
}
#######################
# end of package
#######################
1;
( run in 0.875 second using v1.01-cache-2.11-cpan-39bf76dae61 )