HTML5lib is Back.
Synced with latest version of HTML5lib, which fixes problem with Astral plane characters. I should really do some tests, but the HTML5lib Sanitizer seems to be 2-5 times slower than the old sanitizer.
This commit is contained in:
Jacques Distler
parent
e1a6827f1f
commit
4dd70af5ae
39 changed files with 4843 additions and 5576 deletions
144
vendor/plugins/HTML5lib/tests/test_lxp.rb
vendored
144
vendor/plugins/HTML5lib/tests/test_lxp.rb
vendored
|
|
@ -6,19 +6,19 @@ XMLELEM = /<(\w+\s*)((?:[-:\w]+="[^"]*"\s*)+)(\/?)>/
|
|||
SORTATTRS = '<#{$1+$2.split.sort.join(' ')+$3}>'
|
||||
|
||||
def assert_xml_equal(input, expected=nil, parser=HTML5lib::XMLParser)
|
||||
document = parser.parse(input.chomp).root
|
||||
if not expected
|
||||
expected = input.chomp.gsub(XMLELEM,SORTATTRS)
|
||||
expected = expected.gsub(/&#(\d+);/) {[$1.to_i].pack('U')}
|
||||
output = document.to_s.gsub(/'/,'"').gsub(XMLELEM,SORTATTRS)
|
||||
assert_equal(expected, output)
|
||||
else
|
||||
assert_equal(expected, document.to_s.gsub(/'/,'"'))
|
||||
end
|
||||
document = parser.parse(input.chomp).root
|
||||
if not expected
|
||||
expected = input.chomp.gsub(XMLELEM,SORTATTRS)
|
||||
expected = expected.gsub(/&#(\d+);/) {[$1.to_i].pack('U')}
|
||||
output = document.to_s.gsub(/'/,'"').gsub(XMLELEM,SORTATTRS)
|
||||
assert_equal(expected, output)
|
||||
else
|
||||
assert_equal(expected, document.to_s.gsub(/'/,'"'))
|
||||
end
|
||||
end
|
||||
|
||||
def assert_xhtml_equal(input, expected=nil, parser=HTML5lib::XHTMLParser)
|
||||
assert_xml_equal(input, expected, parser)
|
||||
assert_xml_equal(input, expected, parser)
|
||||
end
|
||||
|
||||
class BasicXhtml5Test < Test::Unit::TestCase
|
||||
|
|
@ -27,8 +27,8 @@ class BasicXhtml5Test < Test::Unit::TestCase
|
|||
assert_xhtml_equal(
|
||||
'<title>Xhtml</title><b><i>content</b></i>',
|
||||
'<html xmlns="http://www.w3.org/1999/xhtml">' +
|
||||
'<head><title>Xhtml</title></head>' +
|
||||
'<body><b><i>content</i></b></body>' +
|
||||
'<head><title>Xhtml</title></head>' +
|
||||
'<body><b><i>content</i></b></body>' +
|
||||
'</html>')
|
||||
end
|
||||
|
||||
|
|
@ -36,8 +36,8 @@ class BasicXhtml5Test < Test::Unit::TestCase
|
|||
assert_xhtml_equal(
|
||||
'<title>mdash</title>A &mdash B',
|
||||
'<html xmlns="http://www.w3.org/1999/xhtml">' +
|
||||
'<head><title>mdash</title></head>' +
|
||||
'<body>A '+ [0x2014].pack('U') + ' B</body>' +
|
||||
'<head><title>mdash</title></head>' +
|
||||
'<body>A '+ [0x2014].pack('U') + ' B</body>' +
|
||||
'</html>')
|
||||
end
|
||||
end
|
||||
|
|
@ -70,24 +70,24 @@ class OpmlTest < Test::Unit::TestCase
|
|||
def test_mixedCaseElement
|
||||
assert_xml_equal(
|
||||
'<opml version="1.0">' +
|
||||
'<head><ownerName>Dave Winer</ownerName></head>' +
|
||||
'<head><ownerName>Dave Winer</ownerName></head>' +
|
||||
'</opml>')
|
||||
end
|
||||
|
||||
def test_mixedCaseAttribute
|
||||
assert_xml_equal(
|
||||
'<opml version="1.0">' +
|
||||
'<body><outline isComment="true"/></body>' +
|
||||
'<body><outline isComment="true"/></body>' +
|
||||
'</opml>')
|
||||
end
|
||||
|
||||
def test_malformed
|
||||
assert_xml_equal(
|
||||
'<opml version="1.0">' +
|
||||
'<body><outline text="Odds & Ends"/></body>' +
|
||||
'<body><outline text="Odds & Ends"/></body>' +
|
||||
'</opml>',
|
||||
'<opml version="1.0">' +
|
||||
'<body><outline text="Odds & Ends"/></body>' +
|
||||
'<body><outline text="Odds & Ends"/></body>' +
|
||||
'</opml>')
|
||||
end
|
||||
end
|
||||
|
|
@ -100,45 +100,45 @@ class XhtmlTest < Test::Unit::TestCase
|
|||
<head><title>MathML</title></head>
|
||||
<body>
|
||||
<math xmlns="http://www.w3.org/1998/Math/MathML">
|
||||
<mrow>
|
||||
<mi>x</mi>
|
||||
<mo>=</mo>
|
||||
|
||||
<mfrac>
|
||||
<mrow>
|
||||
<mi>x</mi>
|
||||
<mo>=</mo>
|
||||
<mrow>
|
||||
<mo>-</mo>
|
||||
<mi>b</mi>
|
||||
</mrow>
|
||||
<mo>±</mo>
|
||||
<msqrt>
|
||||
|
||||
<mfrac>
|
||||
<mrow>
|
||||
<msup>
|
||||
<mi>b</mi>
|
||||
<mn>2</mn>
|
||||
</msup>
|
||||
<mo>-</mo>
|
||||
<mrow>
|
||||
<mrow>
|
||||
<mo>-</mo>
|
||||
<mi>b</mi>
|
||||
</mrow>
|
||||
<mo>±</mo>
|
||||
<msqrt>
|
||||
|
||||
<mrow>
|
||||
<msup>
|
||||
<mi>b</mi>
|
||||
<mn>2</mn>
|
||||
</msup>
|
||||
<mo>-</mo>
|
||||
<mrow>
|
||||
|
||||
<mn>4</mn>
|
||||
<mo>⁢</mo>
|
||||
<mi>a</mi>
|
||||
<mo>⁢</mo>
|
||||
<mi>c</mi>
|
||||
</mrow>
|
||||
</mrow>
|
||||
|
||||
</msqrt>
|
||||
<mn>4</mn>
|
||||
<mo>⁢</mo>
|
||||
<mi>a</mi>
|
||||
<mo>⁢</mo>
|
||||
<mi>c</mi>
|
||||
</mrow>
|
||||
<mrow>
|
||||
<mn>2</mn>
|
||||
<mo>⁢</mo>
|
||||
<mi>a</mi>
|
||||
</mrow>
|
||||
</mfrac>
|
||||
</mrow>
|
||||
|
||||
</msqrt>
|
||||
</mrow>
|
||||
<mrow>
|
||||
<mn>2</mn>
|
||||
<mo>⁢</mo>
|
||||
<mi>a</mi>
|
||||
</mrow>
|
||||
</mfrac>
|
||||
|
||||
</mrow>
|
||||
</math>
|
||||
</body></html>
|
||||
EOX
|
||||
|
|
@ -150,11 +150,11 @@ EOX
|
|||
<head><title>SVG</title></head>
|
||||
<body>
|
||||
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 100 100">
|
||||
<path d="M38,38c0-12,24-15,23-2c0,9-16,13-16,23v7h11v-4c0-9,17-12,17-27
|
||||
c-2-22-45-22-45,3zM45,70h11v11h-11z" fill="#371">
|
||||
</path>
|
||||
<circle cx="50" cy="50" r="45" fill="none" stroke="#371" stroke-width="10">
|
||||
</circle>
|
||||
<path d="M38,38c0-12,24-15,23-2c0,9-16,13-16,23v7h11v-4c0-9,17-12,17-27
|
||||
c-2-22-45-22-45,3zM45,70h11v11h-11z" fill="#371">
|
||||
</path>
|
||||
<circle cx="50" cy="50" r="45" fill="none" stroke="#371" stroke-width="10">
|
||||
</circle>
|
||||
|
||||
</svg>
|
||||
</body></html>
|
||||
|
|
@ -167,24 +167,24 @@ EOX
|
|||
<head><title>XLINK</title></head>
|
||||
<body>
|
||||
<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 100 100">
|
||||
<defs xmlns:l="http://www.w3.org/1999/xlink">
|
||||
<radialGradient id="s1" fx=".4" fy=".2" r=".7">
|
||||
<stop stop-color="#FE8"/>
|
||||
<stop stop-color="#D70" offset="1"/>
|
||||
</radialGradient>
|
||||
<radialGradient id="s2" fx=".8" fy=".5" l:href="#s1"/>
|
||||
<radialGradient id="s3" fx=".5" fy=".9" l:href="#s1"/>
|
||||
<radialGradient id="s4" fx=".1" fy=".5" l:href="#s1"/>
|
||||
</defs>
|
||||
<g stroke="#940">
|
||||
<path d="M73,29c-37-40-62-24-52,4l6-7c-8-16,7-26,42,9z" fill="url(#s1)"/>
|
||||
<path d="M47,8c33-16,48,21,9,47l-6-5c38-27,20-44,5-37z" fill="url(#s2)"/>
|
||||
<path d="M77,32c22,30,10,57-39,51l-1-8c3,3,67,5,36-36z" fill="url(#s3)"/>
|
||||
<defs xmlns:l="http://www.w3.org/1999/xlink">
|
||||
<radialGradient id="s1" fx=".4" fy=".2" r=".7">
|
||||
<stop stop-color="#FE8"/>
|
||||
<stop stop-color="#D70" offset="1"/>
|
||||
</radialGradient>
|
||||
<radialGradient id="s2" fx=".8" fy=".5" l:href="#s1"/>
|
||||
<radialGradient id="s3" fx=".5" fy=".9" l:href="#s1"/>
|
||||
<radialGradient id="s4" fx=".1" fy=".5" l:href="#s1"/>
|
||||
</defs>
|
||||
<g stroke="#940">
|
||||
<path d="M73,29c-37-40-62-24-52,4l6-7c-8-16,7-26,42,9z" fill="url(#s1)"/>
|
||||
<path d="M47,8c33-16,48,21,9,47l-6-5c38-27,20-44,5-37z" fill="url(#s2)"/>
|
||||
<path d="M77,32c22,30,10,57-39,51l-1-8c3,3,67,5,36-36z" fill="url(#s3)"/>
|
||||
|
||||
<path d="M58,84c-4,20-38-4-8-24l-6-5c-36,43,15,56,23,27z" fill="url(#s4)"/>
|
||||
<path d="M40,14c-40,37-37,52-9,68l1-8c-16-13-29-21,16-56z" fill="url(#s1)"/>
|
||||
<path d="M31,33c19,23,20,7,35,41l-9,1.7c-4-19-8-14-31-37z" fill="url(#s2)"/>
|
||||
</g>
|
||||
<path d="M58,84c-4,20-38-4-8-24l-6-5c-36,43,15,56,23,27z" fill="url(#s4)"/>
|
||||
<path d="M40,14c-40,37-37,52-9,68l1-8c-16-13-29-21,16-56z" fill="url(#s1)"/>
|
||||
<path d="M31,33c19,23,20,7,35,41l-9,1.7c-4-19-8-14-31-37z" fill="url(#s2)"/>
|
||||
</g>
|
||||
</svg>
|
||||
</body></html>
|
||||
EOX
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue