Match html5 unquoted attribute parsing

[lhc/web/wiklou.git] / tests / parser / parserTests.txt

diff --git a/tests/parser/parserTests.txt b/tests/parser/parserTests.txt
index 61c331d..9ad7bd5 100644 (file)
--- a/tests/parser/parserTests.txt
+++ b/tests/parser/parserTests.txt
@@ -2520,6 +2520,7 @@ Barack Obama <President> of the United States
 </p>
 !! end
 
+## PHP parser discards the "<pre " string
 !! test
 Handle broken pre-like tags (bug 64025)
 !! options
@@ -2530,13 +2531,8 @@ parsoid=wt2html
 <table><pre </table>
 !! html/php
 <pre>x</pre>
-<table>&lt;pre </table>
+<table><pre></pre></table>
 
-!! html/php+tidy
-<pre>
-x
-</pre>
-<p>&lt;pre</p>
 !! html/parsoid
 <pre about="#mwt1" typeof="mw:Transclusion" data-parsoid='{"a":{"&lt;pre":null},"sa":{"&lt;pre":""},"stx":"html","pi":[[{"k":"1","spc":["","","",""]}]]}' data-mw='{"parts":[{"template":{"target":{"wt":"echo","href":"./Template:Echo"},"params":{"1":{"wt":"&lt;pre &lt;pre>x&lt;/pre>"}},"i":0}}]}'>x</pre>
 
@@ -6334,6 +6330,24 @@ parsoid=wt2html,html2html
 <td data-parsoid='{"startTagSrc":"| ","attrSepSrc":"|","autoInsertedEnd":true}'><a rel="mw:ExtLink" href="ftp://|x||"></a>" onmouseover="alert(document.cookie)">test</td></tr></tbody></table>
 !! end
 
+!! test
+! and || in element attributes should not be parsed as <th>/<td>
+!! wikitext
+{|
+| <div style="color: red !important;" data-contrived="put this here ||">hi</div>
+|}
+!! html/php
+<table>
+<tr>
+<td> <div style="color: red !important;" data-contrived="put this here &#124;&#124;">hi</div>
+</td></tr></table>
+
+!! html/parsoid
+<table>
+<tbody><tr><td> <div style="color: red !important;" data-contrived="put this here ||" data-parsoid='{"stx":"html"}'>hi</div></td></tr>
+</tbody></table>
+!! end
+
 # FIXME: The output seems broken. Filed as T110268.
 !! test
 ! and || in td attributes should not be parsed as <th>/<td>
@@ -6538,14 +6552,18 @@ Table with empty line following the start tag
 
 !! end
 
-# FIXME: Preserve the attribute properly (with an empty string as value) in
-# the PHP parser. Parsoid implements the behavior below.
 !! test
 Table attributes with empty value
 !! wikitext
 {|
 | style=| hello
 |}
+!! html/php
+<table>
+<tr>
+<td style=""> hello
+</td></tr></table>
+
 !! html/parsoid
 <table>
 <tbody>
@@ -8787,6 +8805,40 @@ Mixed list
 
 !! end
 
+!! test
+1. Nested mixed wikitext and html list
+!! wikitext
+* hi
+* <ul><li>ho</li></ul>
+* hi
+** ho
+!! html/php
+<ul><li> hi</li>
+<li> <ul><li>ho</li></ul></li>
+<li> hi
+<ul><li> ho</li></ul></li></ul>
+
+!! html/parsoid
+<ul><li> hi</li>
+<li> <ul data-parsoid='{"stx":"html"}'><li data-parsoid='{"stx":"html"}'>ho</li></ul></li>
+<li> hi
+<ul><li> ho</li></ul></li></ul>
+!! end
+
+!! test
+2. Nested mixed wikitext and html list (incompatible)
+!! wikitext
+; hi
+: {{echo|<li>ho</li>}}
+!! html/php
+<dl><dt> hi</dt>
+<dd> <li>ho</li></dd></dl>
+
+!! html/parsoid
+<dl><dt> hi</dt>
+<dd> <li about="#mwt1" typeof="mw:Transclusion" data-parsoid='{"stx":"html","pi":[[{"k":"1","spc":["","","",""]}]]}' data-mw='{"parts":[{"template":{"target":{"wt":"echo","href":"./Template:Echo"},"params":{"1":{"wt":"&lt;li>ho&lt;/li>"}},"i":0}}]}'>ho</li></dd></dl>
+!! end
+
 !! test
 Nested lists 1
 !! wikitext
@@ -10657,6 +10709,43 @@ foo {{''}} baz
 </p>
 !! end
 
+## This test is about making sure Parsoid's data-mw is well formed in the
+## face of multiple templates with intersecting and overlapping ranges.  The
+## wikitext itself is wretched.
+!! test
+Templates with intersecting and overlapping ranges
+!! wikitext
+{|{{echo|
+<p>ha</p>}}
+{|{{echo|
+<p>ho</p>}}
+{{echo|{{!}}hi}}
+|}
+!! html/php+tidy
+<p>ha</p>
+<p>ho</p>
+<table>
+<tr>
+<td></td>
+</tr>
+<tr>
+<td>hi</td>
+</tr>
+</table>
+<table>
+<tr>
+<td></td>
+</tr>
+</table>
+!! html/parsoid
+<p about="#mwt1" typeof="mw:Transclusion" data-parsoid='{"stx":"html","autoInsertedEnd":true,"pi":[[{"k":"1","spc":["","","",""]}],[{"k":"1","spc":["","","",""]}],[{"k":"1","spc":["","","",""]}]],"firstWikitextNode":"table"}' data-mw='{"parts":["{|",{"template":{"target":{"wt":"echo","href":"./Template:Echo"},"params":{"1":{"wt":"\n&lt;p>ha&lt;/p>"}},"i":0}},"\n","{|",{"template":{"target":{"wt":"echo","href":"./Template:Echo"},"params":{"1":{"wt":"\n&lt;p>ho&lt;/p>"}},"i":1}},"\n",{"template":{"target":{"wt":"echo","href":"./Template:Echo"},"params":{"1":{"wt":"{{!}}hi"}},"i":2}},"\n|}"]}'>ha</p><table about="#mwt1" typeof="mw:ExpandedAttrs" data-mw='{"attribs":[[{"txt":"","html":""},{"html":""}]]}'>
+
+</table><p about="#mwt1">ho</p><table about="#mwt1" typeof="mw:ExpandedAttrs" data-mw='{"attribs":[[{"txt":"","html":""},{"html":""}]]}'>
+
+<tbody><tr><td>hi</td></tr>
+</tbody></table>
+!! end
+
 !! article
 Template:MSGNW test
 !! text
@@ -10880,8 +10969,6 @@ Un-closed <includeonly>
 !! wikitext
 <includeonly>
 !! html
-<p>&lt;includeonly&gt;
-</p>
 !! end
 
 ## We used to, but no longer wt2wt this test since the default serializer
@@ -15368,12 +15455,8 @@ div with illegal double attributes
 
 !!end
 
-# FIXME: produce empty string instead of "class" in the PHP parser, following
-# the HTML5 spec.
 !! test
 div with empty attribute value, space before equals
-!! options
-parsoid
 !! wikitext
 <div class =>HTML rocks</div>
 !! html
@@ -15381,26 +15464,36 @@ parsoid
 
 !! end
 
+# FIXME: Parsoid doesn't match the html5 spec
 !! test
 div with multiple empty attribute values
 !! options
-parsoid
+parsoid=wt2html,html2html
 !! wikitext
 <div id= title=>HTML rocks</div>
-!! html
-<div id="" title="">HTML rocks</div>
+!! html/php
+<div id="title.3D">HTML rocks</div>
 
+!! html/parsoid
+<div id="" title="">HTML rocks</div>
 !! end
 
+# FIXME: Parsoid doesn't match the html5 spec
 !! test
 table with multiple empty attribute values
 !! options
-parsoid
+parsoid=wt2html,html2html
 !! wikitext
 {| title= id=
 | hi
 |}
-!! html
+!! html/php
+<table title="id=">
+<tr>
+<td> hi
+</td></tr></table>
+
+!! html/parsoid
 <table title="" id="">
 <tbody><tr><td> hi</td></tr>
 </tbody></table>
@@ -15417,13 +15510,6 @@ div with braces in attribute value
 <div title="{}">Foo</div>
 !! end
 
-# This it very inconsistent in the PHP parser: it returns 
-# class="class" if there is a space between the name and the equal sign (see
-# 'div with empty attribute value, space before equals'), but strips the
-# attribute completely if the space is missing. We hope that not much content
-# depends on this, so are implementing the behavior below in Parsoid for
-# consistencies' sake.
-# FIXME: fix this behavior in the PHP parser?
 !! test
 div with empty attribute value, no space before equals
 !! options
@@ -15431,7 +15517,7 @@ parsoid=wt2html,html2html
 !! wikitext
 <div class=>HTML rocks</div>
 !! html/php
-<div>HTML rocks</div>
+<div class="">HTML rocks</div>
 
 !! html/parsoid
 <div class="">HTML rocks</div>
@@ -15753,7 +15839,7 @@ Attribute test: equals, then nothing
 !! wikitext
 <font color=>foo</font>
 !! html
-<p><font>foo</font>
+<p><font color="">foo</font>
 </p>
 !! end
 
@@ -20689,9 +20775,11 @@ HTML5 data attributes
 Strip reserved data attributes
 !! wikitext
 <div data-mw="foo" data-parsoid="bar" data-mw-someext="baz" data-ok="fred" data-ooui="xyzzy" data-bad:ns="ns">d</div>
-!! html
+!! html/php
 <div data-ok="fred">d</div>
 
+!! html/parsoid
+<div data-x-data-mw="foo" data-x-data-parsoid="bar" data-x-data-mw-someext="baz" data-ok="fred" data-parsoid='{"stx":"html","a":{"data-ooui":null,"data-bad:ns":null},"sa":{"data-ooui":"xyzzy","data-bad:ns":"ns"}}'>d</div>
 !! end
 
 !! test
@@ -23824,14 +23912,15 @@ HTML tag with 'unnecessary' entity encoding in attributes
 
 !! test
 HTML tag with broken attribute value quoting
+!! options
+parsoid=wt2html,html2html
 !! wikitext
 <span title="Hello world>Foo</span>
 !! html/php
 <p><span title="Hello world">Foo</span>
 </p>
 !! html/parsoid
-<p><span title="Hello world">Foo</span>
-</p>
+<p><span title="Hello world">Foo</span></p>
 !! end
 
 !! test
@@ -23849,6 +23938,8 @@ parsoid=wt2html,html2html
 
 !! test
 Table with broken attribute value quoting
+!! options
+parsoid=wt2html,html2html
 !! wikitext
 {|
 | title="Hello world|Foo
@@ -23869,6 +23960,8 @@ Table with broken attribute value quoting
 
 !! test
 Table with broken attribute value quoting on consecutive lines
+!! options
+parsoid=wt2html,html2html
 !! wikitext
 {|
 | title="Hello world|Foo