sub scrubber { #{{{
return $_scrubber if defined $_scrubber;
+ # Only known uri schemes are allowed to avoid all the ways of
+ # embedding javascrpt.
+ # List at http://en.wikipedia.org/wiki/URI_scheme
+ my $uri_schemes=join("|",
+ # IANA registered schemes
+ "http", "https", "ftp", "mailto", "file", "telnet", "gopher",
+ "aaa", "aaas", "acap", "cap", "cid", "crid",
+ "dav", "dict", "dns", "fax", "go", "h323", "im", "imap",
+ "ldap", "mid", "news", "nfs", "nntp", "pop", "pres",
+ "sip", "sips", "snmp", "tel", "urn", "wais", "xmpp",
+ "z39.50r", "z39.50s",
+ # data is a special case. Allow data:text/<image>, but
+ # disallow data:text/javascript and everything else.
+ qr/data:text\/(?:png|gif|jpeg)/,
+ # Selected unofficial schemes
+ "about", "aim", "callto", "cvs", "ed2k", "feed", "fish", "gg",
+ "irc", "ircs", "lastfm", "ldaps", "magnet", "mms",
+ "msnim", "notes", "rsync", "secondlife", "skype", "ssh",
+ "sftp", "sms", "steam", "webcal", "ymsgr",
+ );
+ my $link=qr/^(?:$uri_schemes:|[^:]+$)/i;
+
eval q{use HTML::Scrubber};
error($@) if $@;
# Lists based on http://feedparser.org/docs/html-sanitization.html
}],
default => [undef, { (
map { $_ => 1 } qw{
- abbr accept accept-charset accesskey action
+ abbr accept accept-charset accesskey
align alt axis border cellpadding cellspacing
char charoff charset checked cite class
clear cols colspan color compact coords
datetime dir disabled enctype for frame
- headers height href hreflang hspace id ismap
+ headers height hreflang hspace id ismap
label lang longdesc maxlength media method
multiple name nohref noshade nowrap prompt
readonly rel rev rows rowspan rules scope
- selected shape size span src start summary
+ selected shape size span start summary
tabindex target title type usemap valign
value vspace width
- poster autoplay loopstart loopend end
+ autoplay loopstart loopend end
playcount controls
} ),
"/" => 1, # emit proper <hr /> XHTML
- }],
+ href => $link,
+ src => $link,
+ action => $link,
+ poster => $link,
+ }],
);
return $_scrubber;
} # }}}
#!/usr/bin/perl
use warnings;
use strict;
-use Test::More tests => 16;
+use Test::More tests => 26;
use Encode;
BEGIN { use_ok("IkiWiki"); }
ok(IkiWiki::htmlize("foo", "mdwn", readfile("t/test2.mdwn")),
"this file crashes markdown if it's fed in as decoded utf-8");
-# embedded javascript sanitisation tests
sub gotcha {
my $html=IkiWiki::htmlize("foo", "mdwn", shift);
return $html =~ /GOTCHA/;
"another entity-encoded CSS script test");
ok(!gotcha(q{<script>GOTCHA</script>}),
"script tag");
+ok(!gotcha(q{<form action="javascript:alert('GOTCHA')">foo</form>}),
+ "form action with javascript");
+ok(!gotcha(q{<video poster="javascript:alert('GOTCHA')" href="foo.avi">foo</video>}),
+ "video poster with javascript");
ok(!gotcha(q{<span style="background: url(javascript:window.location=GOTCHA)">a</span>}),
"CSS script test");
+ok(! gotcha(q{<img src="data:text/javascript:GOTCHA">}),
+ "data:text/javascript (jeez!)");
+ok(gotcha(q{<img src="data:text/png:GOTCHA">}), "data:text/png");
+ok(gotcha(q{<img src="data:text/gif:GOTCHA">}), "data:text/gif");
+ok(gotcha(q{<img src="data:text/jpeg:GOTCHA">}), "data:text/jpeg");
ok(gotcha(q{<p>javascript:alert('GOTCHA')</p>}),
"not javascript AFAIK (but perhaps some web browser would like to
be perverse and assume it is?)");
ok(gotcha(q{<img src="javascript.png?GOTCHA">}), "not javascript");
ok(gotcha(q{<a href="javascript.png?GOTCHA">foo</a>}), "not javascript");
+is(IkiWiki::htmlize("foo", "mdwn",
+ q{<img alt="foo" src="foo.gif">}),
+ q{<img alt="foo" src="foo.gif">}, "img with alt tag allowed");
+is(IkiWiki::htmlize("foo", "mdwn",
+ q{<a href="http://google.com/">}),
+ q{<a href="http://google.com/">}, "absolute url allowed");
+is(IkiWiki::htmlize("foo", "mdwn",
+ q{<a href="foo.html">}),
+ q{<a href="foo.html">}, "relative url allowed");
+is(IkiWiki::htmlize("foo", "mdwn",
+ q{<span class="foo">bar</span>}),
+ q{<span class="foo">bar</span>}, "class attribute allowed");