From: Joey Hess <joey@kodama.kitenet.net>
Date: Sun, 10 Feb 2008 18:22:25 +0000 (-0500)
Subject: cherry-pick uri security fix
X-Git-Tag: 2.31.1~1
X-Git-Url: http://git.vanrenterghem.biz/git.ikiwiki.info.git/commitdiff_plain/df512e55dfbbb912cdf4aee5db48622301bb29ac?hp=-c
cherry-pick uri security fix
---
df512e55dfbbb912cdf4aee5db48622301bb29ac
diff --git a/IkiWiki/Plugin/htmlscrubber.pm b/IkiWiki/Plugin/htmlscrubber.pm
index bc613f924..25caa8a50 100644
--- a/IkiWiki/Plugin/htmlscrubber.pm
+++ b/IkiWiki/Plugin/htmlscrubber.pm
@@ -18,6 +18,28 @@ my $_scrubber;
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
@@ -35,23 +57,27 @@ sub scrubber { #{{{
}],
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;
} # }}}
diff --git a/debian/changelog b/debian/changelog
index b21cdb441..8683bbc1f 100644
--- a/debian/changelog
+++ b/debian/changelog
@@ -1,3 +1,10 @@
+ikiwiki (2.31.1) unstable; urgency=low
+
+ * htmlscrubber security fix: Block javascript in uris.
+ * Add htmlscrubber test suite.
+
+ -- Joey Hess <joeyh@debian.org> Sun, 10 Feb 2008 13:21:35 -0500
+
ikiwiki (2.31) unstable; urgency=low
[ Joey Hess ]
diff --git a/doc/plugins/htmlscrubber.mdwn b/doc/plugins/htmlscrubber.mdwn
index 6ce297a86..d7bcf8099 100644
--- a/doc/plugins/htmlscrubber.mdwn
+++ b/doc/plugins/htmlscrubber.mdwn
@@ -36,3 +36,4 @@ plugin is active:
* <span style="background: url(javascript:window.location='http://example.org/')">CSS script test</span>
* <span style="any: expression(window.location='http://example.org/')">entity-encoded CSS script test</span>
* <span style="any: expression(window.location='http://example.org/')">entity-encoded CSS script test</span>
+* <a href="javascript:alert('foo')">click me</a>
diff --git a/t/htmlize.t b/t/htmlize.t
index a9ccfedcb..edf357010 100755
--- a/t/htmlize.t
+++ b/t/htmlize.t
@@ -1,7 +1,7 @@
#!/usr/bin/perl
use warnings;
use strict;
-use Test::More tests => 4;
+use Test::More tests => 26;
use Encode;
BEGIN { use_ok("IkiWiki"); }
@@ -19,3 +19,52 @@ is(IkiWiki::htmlize("foo", "mdwn", readfile("t/test1.mdwn")),
"utf8; bug #373203");
ok(IkiWiki::htmlize("foo", "mdwn", readfile("t/test2.mdwn")),
"this file crashes markdown if it's fed in as decoded utf-8");
+
+sub gotcha {
+ my $html=IkiWiki::htmlize("foo", "mdwn", shift);
+ return $html =~ /GOTCHA/;
+}
+ok(!gotcha(q{<a href="javascript:alert('GOTCHA')">click me</a>}),
+ "javascript url");
+ok(!gotcha(q{<a href="javascript:alert('GOTCHA')">click me</a>}),
+ "partially encoded javascript url");
+ok(!gotcha(q{<a href="jscript:alert('GOTCHA')">click me</a>}),
+ "jscript url");
+ok(!gotcha(q{<a href="vbscript:alert('GOTCHA')">click me</a>}),
+ "vbscrpt url");
+ok(!gotcha(q{<a href="java script:alert('GOTCHA')">click me</a>}),
+ "java-tab-script url");
+ok(!gotcha(q{<span style="any: expressio(GOTCHA)n(window.location='http://example.org/')">foo</span>}),
+ "entity-encoded CSS script test");
+ok(!gotcha(q{<span style="any: expression(GOTCHA)(window.location='http://example.org/')">foo</span>}),
+ "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");