From 22a018d082ba7ce4f88a44bb473180b5dd58277a Mon Sep 17 00:00:00 2001
From: Aleksander Machniak <alec@alec.pl>
Date: Mon, 18 Jan 2016 03:47:52 -0500
Subject: [PATCH] Merge branch 'dev-svg'
---
program/lib/Roundcube/rcube_washtml.php | 266 +++++++++++++++++++++++++++++++++----------
program/steps/mail/get.inc | 20 ++-
tests/Framework/Washtml.php | 39 ++++++
3 files changed, 256 insertions(+), 69 deletions(-)
diff --git a/program/lib/Roundcube/rcube_washtml.php b/program/lib/Roundcube/rcube_washtml.php
index 1b39b58a..0130a4f 100644
--- a/program/lib/Roundcube/rcube_washtml.php
+++ b/program/lib/Roundcube/rcube_washtml.php
@@ -97,7 +97,20 @@
'tbody', 'td', 'tfoot', 'th', 'thead', 'tr', 'tt', 'u', 'ul', 'var', 'wbr', 'img',
'video', 'source',
// form elements
- 'button', 'input', 'textarea', 'select', 'option', 'optgroup'
+ 'button', 'input', 'textarea', 'select', 'option', 'optgroup',
+ // SVG
+ 'svg', 'altglyph', 'altglyphdef', 'altglyphitem', 'animate',
+ 'animatecolor', 'animatetransform', 'circle', 'clippath', 'defs', 'desc',
+ 'ellipse', 'font', 'g', 'glyph', 'glyphref', 'hkern', 'image', 'line',
+ 'lineargradient', 'marker', 'mask', 'mpath', 'path', 'pattern',
+ 'polygon', 'polyline', 'radialgradient', 'rect', 'set', 'stop', 'switch', 'symbol',
+ 'text', 'textpath', 'tref', 'tspan', 'use', 'view', 'vkern', 'filter',
+ // SVG Filters
+ 'feblend', 'fecolormatrix', 'fecomponenttransfer', 'fecomposite',
+ 'feconvolvematrix', 'fediffuselighting', 'fedisplacementmap',
+ 'feflood', 'fefunca', 'fefuncb', 'fefuncg', 'fefuncr', 'fegaussianblur',
+ 'feimage', 'femerge', 'femergenode', 'femorphology', 'feoffset',
+ 'fespecularlighting', 'fetile', 'feturbulence',
);
/* Ignore these HTML tags and their content */
@@ -110,13 +123,41 @@
'bordercolordark', 'face', 'marginwidth', 'marginheight', 'axis', 'border',
'abbr', 'char', 'charoff', 'clear', 'compact', 'coords', 'vspace', 'hspace',
'cellborder', 'size', 'lang', 'dir', 'usemap', 'shape', 'media',
+ 'background', 'src', 'poster', 'href',
// attributes of form elements
- 'type', 'rows', 'cols', 'disabled', 'readonly', 'checked', 'multiple', 'value'
+ 'type', 'rows', 'cols', 'disabled', 'readonly', 'checked', 'multiple', 'value',
+ // SVG
+ 'accent-height', 'accumulate', 'additive', 'alignment-baseline', 'alphabetic',
+ 'ascent', 'attributename', 'attributetype', 'azimuth', 'basefrequency', 'baseprofile',
+ 'baseline-shift', 'begin', 'bias', 'by', 'clip', 'clip-path', 'clip-rule',
+ 'color', 'color-interpolation', 'color-interpolation-filters', 'color-profile',
+ 'color-rendering', 'cx', 'cy', 'd', 'dx', 'dy', 'diffuseconstant', 'direction',
+ 'display', 'divisor', 'dur', 'edgemode', 'elevation', 'end', 'fill', 'fill-opacity',
+ 'fill-rule', 'filter', 'flood-color', 'flood-opacity', 'font-family', 'font-size',
+ 'font-size-adjust', 'font-stretch', 'font-style', 'font-variant', 'font-weight', 'from',
+ 'fx', 'fy', 'g1', 'g2', 'glyph-name', 'glyphref', 'gradientunits', 'gradienttransform',
+ 'image-rendering', 'in', 'in2', 'k', 'k1', 'k2', 'k3', 'k4', 'kerning', 'keypoints',
+ 'keysplines', 'keytimes', 'lengthadjust', 'letter-spacing', 'kernelmatrix',
+ 'kernelunitlength', 'lighting-color', 'local', 'marker-end', 'marker-mid',
+ 'marker-start', 'markerheight', 'markerunits', 'markerwidth', 'maskcontentunits',
+ 'maskunits', 'max', 'mask', 'mode', 'min', 'numoctaves', 'offset', 'operator',
+ 'opacity', 'order', 'orient', 'orientation', 'origin', 'overflow', 'paint-order',
+ 'path', 'pathlength', 'patterncontentunits', 'patterntransform', 'patternunits',
+ 'points', 'preservealpha', 'r', 'rx', 'ry', 'radius', 'refx', 'refy', 'repeatcount',
+ 'repeatdur', 'restart', 'rotate', 'scale', 'seed', 'shape-rendering', 'show', 'specularconstant',
+ 'specularexponent', 'spreadmethod', 'stddeviation', 'stitchtiles', 'stop-color',
+ 'stop-opacity', 'stroke-dasharray', 'stroke-dashoffset', 'stroke-linecap',
+ 'stroke-linejoin', 'stroke-miterlimit', 'stroke-opacity', 'stroke', 'stroke-width',
+ 'surfacescale', 'targetx', 'targety', 'transform', 'text-anchor', 'text-decoration',
+ 'text-rendering', 'textlength', 'to', 'u1', 'u2', 'unicode', 'values', 'viewbox',
+ 'visibility', 'vert-adv-y', 'version', 'vert-origin-x', 'vert-origin-y', 'word-spacing',
+ 'wrap', 'writing-mode', 'xchannelselector', 'ychannelselector', 'x', 'x1', 'x2',
+ 'xmlns', 'y', 'y1', 'y2', 'z', 'zoomandpan',
);
/* Elements which could be empty and be returned in short form (<tag />) */
static $void_elements = array('area', 'base', 'br', 'col', 'command', 'embed', 'hr',
- 'img', 'input', 'keygen', 'link', 'meta', 'param', 'source', 'track', 'wbr'
+ 'img', 'input', 'keygen', 'link', 'meta', 'param', 'source', 'track', 'wbr',
);
/* State for linked objects in HTML */
@@ -143,13 +184,15 @@
/* Max nesting level */
private $max_nesting_level;
+ private $is_xml = false;
+
/**
* Class constructor
*/
public function __construct($p = array())
{
- $this->_html_elements = array_flip((array)$p['html_elements']) + array_flip(self::$html_elements) ;
+ $this->_html_elements = array_flip((array)$p['html_elements']) + array_flip(self::$html_elements);
$this->_html_attribs = array_flip((array)$p['html_attribs']) + array_flip(self::$html_attribs);
$this->_ignore_elements = array_flip((array)$p['ignore_elements']) + array_flip(self::$ignore_elements);
$this->_void_elements = array_flip((array)$p['void_elements']) + array_flip(self::$void_elements);
@@ -186,22 +229,8 @@
foreach ($this->explode_style($str) as $val) {
if (preg_match('/^url\(/i', $val)) {
if (preg_match('/^url\(\s*[\'"]?([^\'"\)]*)[\'"]?\s*\)/iu', $val, $match)) {
- $url = $match[1];
- if (($src = $this->config['cid_map'][$url])
- || ($src = $this->config['cid_map'][$this->config['base_url'].$url])
- ) {
- $value .= ' url('.htmlspecialchars($src, ENT_QUOTES) . ')';
- }
- else if (preg_match('!^(https?:)?//[a-z0-9/._+-]+$!i', $url, $m)) {
- if ($this->config['allow_remote']) {
- $value .= ' url('.htmlspecialchars($m[0], ENT_QUOTES).')';
- }
- else {
- $this->extlinks = true;
- }
- }
- else if (preg_match('/^data:.+/i', $url)) { // RFC2397
- $value .= ' url('.htmlspecialchars($url, ENT_QUOTES).')';
+ if ($url = $this->wash_uri($match[1])) {
+ $value .= ' url(' . htmlspecialchars($url, ENT_QUOTES) . ')';
}
}
}
@@ -232,54 +261,137 @@
*/
private function wash_attribs($node)
{
- $t = '';
- $washed = '';
+ $result = '';
+ $washed = array();
- foreach ($node->attributes as $key => $plop) {
- $key = strtolower($key);
- $value = $node->getAttribute($key);
+ foreach ($node->attributes as $name => $attr) {
+ $key = strtolower($name);
+ $value = $attr->nodeValue;
- if (isset($this->_html_attribs[$key]) ||
- ($key == 'href' && ($value = trim($value))
- && !preg_match('!^(javascript|vbscript|data:text)!i', $value)
- && preg_match('!^([a-z][a-z0-9.+-]+:|//|#).+!i', $value))
- ) {
- $t .= ' ' . $key . '="' . htmlspecialchars($value, ENT_QUOTES) . '"';
- }
- else if ($key == 'style' && ($style = $this->wash_style($value))) {
+ if ($key == 'style' && ($style = $this->wash_style($value))) {
// replace double quotes to prevent syntax error and XSS issues (#1490227)
- $t .= ' style="' . str_replace('"', '"', $style) . '"';
+ $result .= ' style="' . str_replace('"', '"', $style) . '"';
}
- else if ($key == 'background'
- || ($key == 'src' && preg_match('/^(img|source)$/i', $node->tagName))
- || ($key == 'poster' && strtolower($node->tagName) == 'video')
- ) {
- if (($src = $this->config['cid_map'][$value])
- || ($src = $this->config['cid_map'][$this->config['base_url'].$value])
- ) {
- $t .= ' ' . $key . '="' . htmlspecialchars($src, ENT_QUOTES) . '"';
- }
- else if (preg_match('/^(http|https|ftp):.+/i', $value)) {
- if ($this->config['allow_remote']) {
- $t .= ' ' . $key . '="' . htmlspecialchars($value, ENT_QUOTES) . '"';
+ else if (isset($this->_html_attribs[$key])) {
+ $value = trim($value);
+ $out = null;
+
+ // in SVG to/from attribs may contain anything, including URIs
+ if ($key == 'to' || $key == 'from') {
+ $key = strtolower($node->getAttribute('attributeName'));
+ if ($key && !isset($this->_html_attribs[$key])) {
+ $key = null;
}
- else {
- $this->extlinks = true;
- if ($this->config['blocked_src']) {
- $t .= ' ' . $key . '="' . htmlspecialchars($this->config['blocked_src'], ENT_QUOTES) . '"';
+ }
+
+ if ($this->is_image_attribute($node->tagName, $key)) {
+ $out = $this->wash_uri($value, true);
+ }
+ else if ($this->is_link_attribute($node->tagName, $key)) {
+ if (!preg_match('!^(javascript|vbscript|data:text)!i', $value)
+ && preg_match('!^([a-z][a-z0-9.+-]+:|//|#).+!i', $value)
+ ) {
+ $out = $value;
+ }
+ }
+ else if ($this->is_funciri_attribute($node->tagName, $key)) {
+ if (preg_match('/^[a-z:]*url\(/i', $val)) {
+ if (preg_match('/^([a-z:]*url)\(\s*[\'"]?([^\'"\)]*)[\'"]?\s*\)/iu', $value, $match)) {
+ if ($url = $this->wash_uri($match[2])) {
+ $result .= ' ' . $attr->nodeName . '="' . $match[1] . '(' . htmlspecialchars($url, ENT_QUOTES) . ')'
+ . substr($val, strlen($match[0])) . '"';
+ continue;
+ }
+ }
+ else {
+ $out = $value;
}
}
+ else {
+ $out = $value;
+ }
}
- else if (preg_match('/^data:.+/i', $value)) { // RFC2397
- $t .= ' ' . $key . '="' . htmlspecialchars($value, ENT_QUOTES) . '"';
+ else if ($key) {
+ $out = $value;
+ }
+
+ if ($out !== null && $out !== '') {
+ $result .= ' ' . $attr->nodeName . '="' . htmlspecialchars($out, ENT_QUOTES) . '"';
+ }
+ else if ($value) {
+ $washed[] = htmlspecialchars($attr->nodeName, ENT_QUOTES);
}
}
else {
- $washed .= ($washed ? ' ' : '') . $key;
+ $washed[] = htmlspecialchars($attr->nodeName, ENT_QUOTES);
}
}
- return $t . ($washed && $this->config['show_washed'] ? ' x-washed="'.$washed.'"' : '');
+ if (!empty($washed) && $this->config['show_washed']) {
+ $result .= ' x-washed="' . implode(' ', $washed) . '"';
+ }
+
+ return $result;
+ }
+
+ /**
+ * Wash URI value
+ */
+ private function wash_uri($uri, $blocked_source = false)
+ {
+ if (($src = $this->config['cid_map'][$uri])
+ || ($src = $this->config['cid_map'][$this->config['base_url'].$uri])
+ ) {
+ return $src;
+ }
+
+ // allow url(#id) used in SVG
+ if ($uri[0] == '#') {
+ return $uri;
+ }
+
+ if (preg_match('/^(http|https|ftp):.+/i', $uri)) {
+ if ($this->config['allow_remote']) {
+ return $uri;
+ }
+
+ $this->extlinks = true;
+ if ($blocked_source && $this->config['blocked_src']) {
+ return $this->config['blocked_src'];
+ }
+ }
+ else if (preg_match('/^data:image.+/i', $uri)) { // RFC2397
+ return $uri;
+ }
+ }
+
+ /**
+ * Check it the tag/attribute may contain an URI
+ */
+ private function is_link_attribute($tag, $attr)
+ {
+ return $tag == 'a' && $attr == 'href';
+ }
+
+ /**
+ * Check it the tag/attribute may contain an image URI
+ */
+ private function is_image_attribute($tag, $attr)
+ {
+ return $attr == 'background'
+ || $attr == 'color-profile' // SVG
+ || ($attr == 'poster' && $tag == 'video')
+ || ($attr == 'src' && preg_match('/^(img|source)$/i', $tag))
+ || ($tag == 'image' && $attr == 'href'); // SVG
+ }
+
+ /**
+ * Check it the tag/attribute may contain a FUNCIRI value
+ */
+ private function is_funciri_attribute($tag, $attr)
+ {
+ return in_array($attr, array('fill', 'filter', 'stroke', 'marker-start',
+ 'marker-end', 'marker-mid', 'clip-path', 'mask', 'cursor'));
}
/**
@@ -322,14 +434,31 @@
}
else if (isset($this->_html_elements[$tagName])) {
$content = $this->dumpHtml($node, $level);
- $dump .= '<' . $tagName . $this->wash_attribs($node) .
- ($content === '' && isset($this->_void_elements[$tagName]) ? ' />' : ">$content</$tagName>");
+ $dump .= '<' . $node->tagName;
+
+ if ($tagName == 'svg') {
+ $xpath = new DOMXPath($node->ownerDocument);
+ foreach ($xpath->query('namespace::*') as $ns) {
+ if ($ns->nodeName != 'xmlns:xml') {
+ $dump .= ' ' . $ns->nodeName . '="' . $ns->nodeValue . '"';
+ }
+ }
+ }
+
+ $dump .= $this->wash_attribs($node);
+
+ if ($content === '' && ($this->is_xml || isset($this->_void_elements[$tagName]))) {
+ $dump .= ' />';
+ }
+ else {
+ $dump .= '>' . $content . '</' . $node->tagName . '>';
+ }
}
else if (isset($this->_ignore_elements[$tagName])) {
- $dump .= '<!-- ' . htmlspecialchars($tagName, ENT_QUOTES) . ' not allowed -->';
+ $dump .= '<!-- ' . htmlspecialchars($node->tagName, ENT_QUOTES) . ' not allowed -->';
}
else {
- $dump .= '<!-- ' . htmlspecialchars($tagName, ENT_QUOTES) . ' ignored -->';
+ $dump .= '<!-- ' . htmlspecialchars($node->tagName, ENT_QUOTES) . ' ignored -->';
$dump .= $this->dumpHtml($node, $level); // ignore tags not its content
}
break;
@@ -375,12 +504,18 @@
// Detect max nesting level (for dumpHTML) (#1489110)
$this->max_nesting_level = (int) @ini_get('xdebug.max_nesting_level');
+ // SVG need to be parsed as XML
+ $this->is_xml = stripos($html, '<svg') !== false || stripos($html, '<?xml') !== false;
+ $method = $this->is_xml ? 'loadXML' : 'loadHTML';
+ $options = 0;
+
// Use optimizations if supported
if (PHP_VERSION_ID >= 50400) {
- @$node->loadHTML($html, LIBXML_PARSEHUGE | LIBXML_COMPACT);
+ $options = LIBXML_PARSEHUGE | LIBXML_COMPACT | LIBXML_NONET;
+ @$node->{$method}($html, $options);
}
else {
- @$node->loadHTML($html);
+ @$node->{$method}($html);
}
return $this->dumpHtml($node);
@@ -399,6 +534,8 @@
*/
private function cleanup($html)
{
+ $html = trim($html);
+
// special replacements (not properly handled by washtml class)
$html_search = array(
// space(s) between <NOBR>
@@ -420,17 +557,19 @@
'',
'<html>',
);
+
$html = preg_replace($html_search, $html_replace, trim($html));
- //-> Replace all of those weird MS Word quotes and other high characters
+ // Replace all of those weird MS Word quotes and other high characters
$badwordchars = array(
"\xe2\x80\x98", // left single quote
"\xe2\x80\x99", // right single quote
"\xe2\x80\x9c", // left double quote
"\xe2\x80\x9d", // right double quote
"\xe2\x80\x94", // em dash
- "\xe2\x80\xa6" // elipses
+ "\xe2\x80\xa6" // elipses
);
+
$fixedwordchars = array(
"'",
"'",
@@ -439,6 +578,7 @@
'—',
'...'
);
+
$html = str_replace($badwordchars, $fixedwordchars, $html);
// PCRE errors handling (#1486856), should we use something like for every preg_* use?
@@ -484,7 +624,7 @@
$tagname = $matches[2];
$tagname = preg_replace(array(
'/:.*$/', // Microsoft's Smart Tags <st1:xxxx>
- '/[^a-z0-9_\[\]\!-]/i', // forbidden characters
+ '/[^a-z0-9_\[\]\!?-]/i', // forbidden characters
), '', $tagname);
// fix invalid closing tags - remove any attributes (#1489446)
diff --git a/program/steps/mail/get.inc b/program/steps/mail/get.inc
index 96cdd77..6a70315 100644
--- a/program/steps/mail/get.inc
+++ b/program/steps/mail/get.inc
@@ -515,12 +515,20 @@
*/
function rcmail_svg_filter($body)
{
- $dom = new DOMDocument;
- $dom->loadXML($body);
+ // clean SVG with washhtml
+ $wash_opts = array(
+ 'show_washed' => false,
+ 'allow_remote' => false,
+ 'charset' => RCUBE_CHARSET,
+ 'html_elements' => array('title'),
+// 'blocked_src' => 'program/resources/blocked.gif',
+ );
- foreach ($dom->getElementsByTagName('script') as $node) {
- $node->parentNode->removeChild($node);
- }
+ // initialize HTML washer
+ $washer = new rcube_washtml($wash_opts);
- return $dom->saveXML() ?: '';
+ // allow CSS styles, will be sanitized by rcmail_washtml_callback()
+ $washer->add_callback('style', 'rcmail_washtml_callback');
+
+ return $washer->wash($body);
}
diff --git a/tests/Framework/Washtml.php b/tests/Framework/Washtml.php
index 5903d0c..9515f0d 100644
--- a/tests/Framework/Washtml.php
+++ b/tests/Framework/Washtml.php
@@ -213,4 +213,43 @@
$this->assertTrue(strpos($washed, $exp) !== false, "Style quotes XSS issue (#1490227)");
}
+
+ /**
+ * Test SVG cleanup
+ */
+ function test_style_wash_svg()
+ {
+ $svg = '<?xml version="1.0" standalone="no"?>
+<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd">
+<svg version="1.1" baseProfile="full" xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:cc="http://creativecommons.org/ns#" viewBox="0 0 100 100">
+ <polygon id="triangle" points="0,0 0,50 50,0" fill="#009900" stroke="#004400" onmouseover="alert(1)" />
+ <text x="50" y="68" font-size="48" fill="#FFF" text-anchor="middle"><![CDATA[410]]></text>
+ <script type="text/javascript">
+ alert(document.cookie);
+ </script>
+ <text x="10" y="25" >An example text</text>
+ <a xlink:href="http://www.w.pl"><rect width="100%" height="100%" /></a>
+ <foreignObject xlink:href="data:text/xml,%3Cscript xmlns=\'http://www.w3.org/1999/xhtml\'%3Ealert(1)%3C/script%3E"/>
+ <set attributeName="onmouseover" to="alert(1)"/>
+ <animate attributeName="onunload" to="alert(1)"/>
+ <animate attributeName="xlink:href" begin="0" from="javascript:alert(1)" />
+</svg>';
+
+ $exp = '<svg xmlns:cc="http://creativecommons.org/ns#" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns="http://www.w3.org/2000/svg" version="1.1" baseProfile="full" viewBox="0 0 100 100">
+ <polygon id="triangle" points="0,0 0,50 50,0" fill="#009900" stroke="#004400" x-washed="onmouseover" />
+ <text x="50" y="68" font-size="48" fill="#FFF" text-anchor="middle">410</text>
+ <!-- script not allowed -->
+ <text x="10" y="25">An example text</text>
+ <a xlink:href="http://www.w.pl"><rect width="100%" height="100%" /></a>
+ <!-- foreignObject ignored -->
+ <set attributeName="onmouseover" x-washed="to" />
+ <animate attributeName="onunload" x-washed="to" />
+ <animate attributeName="xlink:href" begin="0" x-washed="from" />
+</svg>';
+
+ $washer = new rcube_washtml;
+ $washed = $washer->wash($svg);
+
+ $this->assertSame($washed, $exp, "SVG content");
+ }
}
--
Gitblit v1.9.1