Merge branch 'master' of ssh://git.ikiwiki.info/srv/git/ikiwiki.info
[ikiwiki.git] / IkiWiki / Plugin / htmlscrubber.pm
bloba58a27d5221acc50e932057b9fabaffcf38c7b97
1 #!/usr/bin/perl
2 package IkiWiki::Plugin::htmlscrubber;
4 use warnings;
5 use strict;
6 use IkiWiki 3.00;
8 # This regexp matches urls that are in a known safe scheme.
9 # Feel free to use it from other plugins.
10 our $safe_url_regexp;
12 sub import {
13 hook(type => "getsetup", id => "htmlscrubber", call => \&getsetup);
14 hook(type => "sanitize", id => "htmlscrubber", call => \&sanitize);
16 # Only known uri schemes are allowed to avoid all the ways of
17 # embedding javascrpt.
18 # List at http://en.wikipedia.org/wiki/URI_scheme
19 my $uri_schemes=join("|", map quotemeta,
20 # IANA registered schemes
21 "http", "https", "ftp", "mailto", "file", "telnet", "gopher",
22 "aaa", "aaas", "acap", "cap", "cid", "crid",
23 "dav", "dict", "dns", "fax", "go", "h323", "im", "imap",
24 "ldap", "mid", "news", "nfs", "nntp", "pop", "pres",
25 "sip", "sips", "snmp", "tel", "urn", "wais", "xmpp",
26 "z39.50r", "z39.50s",
27 # Selected unofficial schemes
28 "aim", "callto", "cvs", "ed2k", "feed", "fish", "gg",
29 "irc", "ircs", "lastfm", "ldaps", "magnet", "mms",
30 "msnim", "notes", "rsync", "secondlife", "skype", "ssh",
31 "sftp", "smb", "sms", "snews", "webcal", "ymsgr",
33 # data is a special case. Allow a few data:image/ types,
34 # but disallow data:text/javascript and everything else.
35 $safe_url_regexp=qr/^(?:(?:$uri_schemes):|data:image\/(?:png|jpeg|gif)|[^:]+(?:$|[\/\?#]))|^#/i;
38 sub getsetup () {
39 return
40 plugin => {
41 safe => 1,
42 rebuild => undef,
43 section => "core",
45 htmlscrubber_skip => {
46 type => "pagespec",
47 example => "!*/Discussion",
48 description => "PageSpec specifying pages not to scrub",
49 link => "ikiwiki/PageSpec",
50 safe => 1,
51 rebuild => undef,
55 sub sanitize (@) {
56 my %params=@_;
58 if (exists $config{htmlscrubber_skip} &&
59 length $config{htmlscrubber_skip} &&
60 exists $params{page} &&
61 pagespec_match($params{page}, $config{htmlscrubber_skip})) {
62 return $params{content};
65 return scrubber()->scrub($params{content});
68 my $_scrubber;
69 sub scrubber {
70 return $_scrubber if defined $_scrubber;
72 eval q{use HTML::Scrubber};
73 error($@) if $@;
74 # Lists based on http://feedparser.org/docs/html-sanitization.html
75 # With html5 tags added.
76 $_scrubber = HTML::Scrubber->new(
77 allow => [qw{
78 a abbr acronym address area b big blockquote br br/
79 button caption center cite code col colgroup dd del
80 dfn dir div dl dt em fieldset font form h1 h2 h3 h4
81 h5 h6 hr hr/ i img input ins kbd label legend li map
82 menu ol optgroup option p p/ pre q s samp select small
83 span strike strong sub sup table tbody td textarea
84 tfoot th thead tr tt u ul var
86 video audio source section nav article aside hgroup
87 header footer figure figcaption time mark canvas
88 datalist progress meter ruby rt rp details summary
89 }],
90 default => [undef, { (
91 map { $_ => 1 } qw{
92 abbr accept accept-charset accesskey
93 align alt axis border cellpadding cellspacing
94 char charoff charset checked class
95 clear cols colspan color compact coords
96 datetime dir disabled enctype for frame
97 headers height hreflang hspace id ismap
98 label lang maxlength media method
99 multiple name nohref noshade nowrap prompt
100 readonly rel rev rows rowspan rules scope
101 selected shape size span start summary
102 tabindex target title type valign
103 value vspace width
105 autofocus autoplay preload loopstart
106 loopend end playcount controls pubdate
107 placeholder min max step low high optimum
108 form required autocomplete novalidate pattern
109 list formenctype formmethod formnovalidate
110 formtarget reversed spellcheck open hidden
111 } ),
112 "/" => 1, # emit proper <hr /> XHTML
113 href => $safe_url_regexp,
114 src => $safe_url_regexp,
115 action => $safe_url_regexp,
116 formaction => $safe_url_regexp,
117 cite => $safe_url_regexp,
118 longdesc => $safe_url_regexp,
119 poster => $safe_url_regexp,
120 usemap => $safe_url_regexp,
123 return $_scrubber;