From acab3167f89b62fdb3a988ad1a6f67537b9fd59c Mon Sep 17 00:00:00 2001 From: Arlo Breault Date: Mon, 16 Jul 2018 18:56:16 -0400 Subject: [PATCH] Sync up with Parsoid parserTests.txt This now aligns with Parsoid commit d473791ea982178af7a0fe15aff5cf8e21aaa5e8 Change-Id: Ide8a0d8df20242c9e60e40db653368ba3ef7a06c --- tests/parser/parserTests.txt | 326 ++++++++++++++++++++--------------- 1 file changed, 185 insertions(+), 141 deletions(-) diff --git a/tests/parser/parserTests.txt b/tests/parser/parserTests.txt index 32542711aa..e2da66282e 100644 --- a/tests/parser/parserTests.txt +++ b/tests/parser/parserTests.txt @@ -1394,7 +1394,7 @@ Non-word characters don't terminate tag names (T19663, T42670, T54022) doesn't terminate -!! html +!! html/php

<blockquote|>a</blockquote>

<b→> doesn't terminate </b→>

<bä> doesn't terminate </bä> @@ -1457,12 +1457,10 @@ parsoid=wt2html s -!! html/php+tidy +!! html+tidy

<s.foo>s

-!! html/parsoid -

<s.foo>s

!! end ### @@ -2021,22 +2019,18 @@ a
foo

!! end -# Remex wraps empty tag runs with p-tags. -# Parsoid strips them out during p-wrapping. !! test No p-wrappable content +!! options +parsoid=wt2html,html2html !! wikitext
x
x
x
 -!! html/php+tidy +!! html+tidy
x
x

x
 -!! html/parsoid -
x
 -
x
 -
x
 !! end # T177612: Parsoid-only test @@ -2052,6 +2046,8 @@ x !! test Block tag on one line (
) +!! options +parsoid=wt2html,html2html !! wikitext a
foo
@@ -2083,6 +2079,8 @@ b

foo
!! test Block tag on both lines (
) +!! options +parsoid=wt2html,html2html !! wikitext a
foo
@@ -2122,12 +2120,6 @@ x

foo

z

!! end -# The difference between Parsoid & Remex here -# is because of Parsoid's Tidy-emulation code -# for p-wrapping. We'll start work to remove this -# emulation code in Parsoid sooner than later. -# Remex wraps empty tag runs with p-tags. -# Parsoid strips them out in a separate pass. !! test Empty lines between lines with block tags !! wikitext @@ -2349,6 +2341,22 @@ hi

!! end +!! test +Dangling table row doesn't prevent p-wrapping +!! options +parsoid=wt2html,html2html +!! wikitext +{| +| hi +|- +|} ho +!! html/parsoid + + + +
hi

ho

+!! end + ### ### Preformatted text ### @@ -2486,6 +2494,8 @@ Bar !! test T17491: / in blockquote +!! options +parsoid=wt2html !! wikitext
Foo bar baz quux @@ -2505,6 +2515,8 @@ Foo bar baz quux !! test T17491: / in blockquote (2) +!! options +parsoid=wt2html !! wikitext
Foo bar baz quux
@@ -2838,6 +2850,7 @@ Barack Obama of the United States

!! end +## Remex doesn't account for fostered content. !! test Handle broken pre-like tags (T66025) !! options @@ -2856,7 +2869,7 @@ parsoid=wt2html !! html/parsoid 
x
-<pre
+

<pre

!! end !! test @@ -3518,8 +3531,10 @@ parsoid=wt2html,html2html !! end -!!test +!! test 3a. Indent-Pre and block tags (single-line html) +!! options +parsoid=wt2html,html2html !! wikitext a

foo

b
foo
@@ -3531,18 +3546,17 @@ parsoid=wt2html,html2html c
foo 
 foo
-!! html/parsoid -

a

foo

-

b

foo
-

c

foo
-
 foo 
-
!! html/php+tidy

a

foo

b

foo

c

foo 

 foo
+!! html/parsoid +

a

foo

+ b

foo

+ c

foo

+
 foo
!! end !! test @@ -3630,17 +3644,17 @@ foo !! test 4. Indent-Pre and extension tags +!! options +parsoid=wt2html,html2html !! wikitext a -!! html/php - a 
-NULL
+!! html/php+tidy
+
 a 
NULL
 array (
 )
 

-
 !! html/parsoid
- a 

+
 a 

 !! end
 
 !! test
@@ -6851,7 +6865,7 @@ parsoid=wt2html,html2html
 
ho">ha</div>
 

 !! html/parsoid
-ho">ha
+
ho">ha

 !! end
 
 !! test
@@ -7313,10 +7327,11 @@ Wikitext table with html-syntax row
 foo
 !! end
 
+## Remex doesn't account for fostered content.
 !! test
 Fostered content in tables: Plain text
 !! options
-parsoid=wt2html,html2html
+parsoid=wt2html
 !! wikitext
 {|
 |-
@@ -7334,9 +7349,9 @@ a
 a
 

 !! html/parsoid
-
-a
-
+

+a

+

 

 !! end
 
@@ -10150,6 +10165,8 @@ Unbalanced closing block tags break a list
 
 !! test
 Unbalanced closing non-block tags don't break a list
+!! options
+parsoid=wt2html,html2html
 !! wikitext
 
 *a
@@ -10160,12 +10177,9 @@ Unbalanced closing non-block tags don't break a list
 
  • a
    • 
 
  • b

 !! html/parsoid
-
-
    
-
  • a
    • 
-
  • b
    • 
-

-
+

+
  • a
    • 
+
  • b

 !! end
 
 # Parsoid does some post-dom-building cleanup
@@ -10186,14 +10200,7 @@ parsoid=wt2html,wt2wt,html2html
 !! end
 
 # Output is ugly because of all the misnested tag fixups.
-# Remex is wrapping p-tags around empty elements.
-# Parsoid has special-case handling of this pattern of
-# wrapping lists in formatting tags.
-# FIXME: Should we remove this code from Parsoid? Or add
-# special support in Remex? If the latter, maybe just wait
-# for Parsoid to become the default parser.
-# See T70395.
-!!test
+!! test
 1. List embedded in a formatting tag
 !! wikitext
 
@@ -10204,17 +10211,13 @@ parsoid=wt2html,wt2wt,html2html
 
  • foo

 

 !! html/parsoid
-
-
    
-
  • foo
    • 
-

-
-!!end
+

+
  • foo

+

+!! end
 
-# Output is ugly because of all the misnested tag fixups
-# Remex is wrapping p-tags around empty elements.
-# Parsoid has code that strips useless p-tags.
-!!test
+# Output is ugly because of all the misnested tag fixups.
+!! test
 2. List embedded in a formatting tag in a misnested way
 !! wikitext
 
@@ -10225,13 +10228,10 @@ parsoid=wt2html,wt2wt,html2html
 
  • a
    • 
 
  • b

 !! html/parsoid
-
-
    
-
  • a
    • 
-    
-
  • b
    • 
-

-!!end
+

+
  • a
    • 
+
  • b

+!! end
 
 ###
 ### Magic Words
@@ -12918,10 +12918,7 @@ Templates: Block Tags: 2. Back-to-back template uses
 
 !!end
 
-## This is an edge case relating to paragraph wrapping.
-## Note that Parsoid fails to match Remex because it's using the closing tag
-## as a heuristic to determine if it's in a block, rather than SAX based events.
-## And Parsoid drops empty elements in templates.
+## Parsoid drops empty elements in templates.
 !! test
 Templates: Correctly encapsulate templates producing 
 tag without a corresponding 
 tag
 !! wikitext
@@ -12933,7 +12930,7 @@ b
}}
 b

 !! html/parsoid
 
a

-b
+
b

 !! end
 
 !!test
@@ -13303,6 +13300,7 @@ parsoid=wt2html,wt2wt
 
foo

 !!end
 
+## Remex doesn't account for fostered content.
 !! test
 Templates: Wiki Tables: 1a. Fostering of entire template content
 !! wikitext
@@ -13319,7 +13317,7 @@ a
 a
 

 !! html/parsoid
-a
+
a

 
 

 !! end
@@ -13355,6 +13353,7 @@ foo
 
 !! end
 
+## Remex doesn't account for fostered content.
 !! test
 Templates: Wiki Tables: 2. Fostering of partial template content
 !! wikitext
@@ -13374,7 +13373,7 @@ a
 
b

 

 !! html/parsoid
-a
b

+
a
b

 
 
 

@@ -13438,20 +13437,20 @@ Templates: Wiki Tables: 6. Templated tags, templated td-tags
 ## a  marker tag for  tags and they are expanded
 ## much later. We are verifying that this  tag usage
 ## doesn't prevent foster parenting.
-!!test
+!! test
 Templates: Wiki Tables: 7. Fosterable s should get fostered
-!!wikitext
+!! wikitext
 {{PartialTable}}foo
 |}
 
 
-!!html/parsoid
-[1]
+!! html/parsoid
+
[1]

 
 

 
-
  1. ↑  foo

-!!end
+
  1. ↑  foo

+!! end
 
 !! test
 Templates: Wiki Tables: 8. Fosterable meta-tags should get fostered
@@ -14579,8 +14578,8 @@ thumbsize=220
 

 !! html/parsoid
 
123456

-
123
456

-
123
456

+
123
456
+123
456

 !! end
 
 !! test
@@ -17730,6 +17729,8 @@ Media link with text
 # Parsoid & Remex fix the p-wrapping since they operate on the DOM.
 !! test
 Media link with nasty text
+!! options
+parsoid=wt2html,html2html
 !! wikitext
 [[Media:Foobar.jpg|Safe Link
" onmouseover="alert(document.cookie)" onfoo="
]]
 !! html/php
@@ -17738,8 +17739,7 @@ Media link with nasty text
 !! html/php+tidy
 
Safe Link
" onmouseover="alert(document.cookie)" onfoo="

 !! html/parsoid
-
Safe Link

-
+
Safe Link
" onmouseover="alert(document.cookie)" onfoo="

 !! end
 
 !! test
@@ -18453,6 +18453,8 @@ array (
 
 !! test
 Parser hook: nested tags
+!! options
+parsoid=wt2html
 !! wikitext
 
 !! html/php
@@ -18462,8 +18464,14 @@ array (
 )
 
</tag>
 
+!! html/php+tidy
+
''
+array (
+)
+
</tag>
+

 !! html/parsoid
-
</tag>
+
</tag>

 !! end
 
 !! test
@@ -18539,7 +18547,7 @@ array (
 text
 
 !! html/parsoid
-
text
+
text

 !! end
 
 ##  should be output literally since there is no matching tag that begins it
@@ -18671,8 +18679,6 @@ Nested template calls
 ### Sanitizer
 ###
 
-# Remex wraps empty tag runs with p-tags.
-# Parsoid strips them out during p-wrapping.
 !! test
 Sanitizer: Closing of open tags
 !! wikitext
@@ -18680,7 +18686,7 @@ Sanitizer: Closing of open tags
 !! html/php+tidy
 

 !! html/parsoid
-

+

 !! end
 
 !! test
@@ -18698,10 +18704,9 @@ Sanitizer: Closing of closed but not open tags
 parsoid=wt2html
 !! wikitext
 
-!! html/php+tidy
+!! html+tidy
 

 

-!! html/parsoid
 !! end
 
 !! test
@@ -19278,7 +19283,8 @@ http://===r:::https://b
 
 !! end
 
-# Known to produce bad XML for now
+## Remex doesn't account for fostered content.
+## Known to produce bad XML for now
 !! test
 Fuzz testing: Parser24
 !! options
@@ -19305,14 +19311,27 @@ MOVE YOUR MOUSE CURSOR OVER THIS TEXT
 
 
 
-!! html/parsoid
-
{{{|
-}}}} >
-
MOVE YOUR MOUSE CURSOR OVER THIS TEXT

+!! html/php+tidy
 
+{{{|
+}}}} >
+

 
+MOVE YOUR MOUSE CURSOR OVER THIS TEXT
+

+
+
+

+

+

+!! html/parsoid
+

+{{{|
+}}}} >
+

 
-
+MOVE YOUR MOUSE CURSOR OVER THIS TEXT

+

 !! end
 
 # Note: the current result listed for this is not what the original one was,
@@ -19504,6 +19523,10 @@ New wiki paragraph
 
Bold paragraph
 
New wiki paragraph
 

+!! html/php+tidy
+
Bold paragraph
+
New wiki paragraph
+

 !! html/parsoid
 
Bold paragraph
 
New wiki paragraph
@@ -20453,7 +20476,7 @@ parsoid=wt2html
 '''''
 !! html/php
 !! html/parsoid
-
+

 !! end
 
 # same html as previous, but wikitext adjusted to match parsoid html2wt
@@ -22807,11 +22830,11 @@ language=zh variant=zh-cn
 ad
 
 !! html/parsoid
-
a
d

+a
d
 
-
a
d

+a
d
 
-
ad

+
ad

 !! end
 
 !! test
@@ -23095,10 +23118,8 @@ language=zh variant=zh-cn
 
-{zh-cn:AAA

 

 !! html/parsoid
-
foo:bar

-
-{zh-cn

-
AAA

-

+
foo:bar

+
-{zh-cn
AAA

 !! end
 
 !! test
@@ -23463,7 +23484,6 @@ Line two
 !! end
 
 # doBlockLevels screws up this output and Remex cleans up as much as it can.
-# Parsoid seems to do a better job here since its p-wrapper is probably smarter.
 !! test
 Nesting tags, paragraphs on lines which begin with 

 !! wikitext
@@ -23474,15 +23494,15 @@ B
 
B
 

 !! html/parsoid
-

-
A
-B
-

+
A

+
B

 !! end
 
 # T8200: 
 should behave like 
 with respect to line breaks
 !! test
 T8200: paragraphs inside blockquotes (no extra line breaks)
+!! options
+parsoid=wt2html,html2html
 !! wikitext
 
Line one
 
@@ -23498,6 +23518,8 @@ Line two

 
 !! test
 T8200: paragraphs inside blockquotes (extra line break on open)
+!! options
+parsoid=wt2html,html2html
 !! wikitext
 

 Line one
@@ -23516,9 +23538,10 @@ Line two

 Line two

 !! end
 
-# Parsoid's output is broken on this because of Tidy-compatibility cruft
 !! test
 T8200: paragraphs inside blockquotes (extra line break on close)
+!! options
+parsoid=wt2html,html2html
 !! wikitext
 
Line one
 
@@ -24514,7 +24537,7 @@ __TOC__
 
Quote
[edit]

 !! html/parsoid
 
-
Quote

+
Quote

 !! end
 
 !! test
@@ -24578,7 +24601,7 @@ __TOC__
 
 
Foo Bar

 
-
Foo 
Bar

+
Foo 
Bar

 !! end
 
 # Don't expect Parsoid to roundtrip this until the php parser comes closer to
@@ -25028,13 +25051,13 @@ parsoid=wt2html,wt2wt
 #### Parsoid-specific functionality tests
 #### -----------------------------------------------------------------
 
-# T65642/T68749: Formatting elt fixup around images is cleaned up.
+# T65642/T68749: Formatting elt fixup around images.
 # We know wt2wt will fail, but we expect selser to pass.
 # Due to the nature of our testing, wt2wt and selser tests will enter the
 # blacklist and we'll catch selser regressions based on changes to the
 # blacklist entries for selser tests.
 !! test
-1. Bad treebuilder fixup of formatting elt is cleaned up
+1. Treebuilder fixup of formatting elt
 !! options
 parsoid=wt2html,wt2wt
 !! wikitext
@@ -25044,50 +25067,63 @@ parsoid=wt2html,wt2wt
 [[Image:Foobar.jpg|right|Test]]
 
 |}
+!! html/php+tidy
+
+
+

+

+

+
Test

+

+

+

 !! html/parsoid
 
+

+
Test

+

 

-
-
Test

-
-

 

 !! end
 
 !! test
-2. Bad treebuilder fixup of formatting elt is cleaned up
+2. Treebuilder fixup of formatting elt
 !! options
 parsoid=wt2html,wt2wt
 !! wikitext
 '''foo[[File:Foobar.jpg|thumb|caption]]bar'''
 
 [[Image:Foobar.jpg|right|300px]]
+!! html/php+tidy
+
foo
  
caption
bar
+
Foobar.jpg

 !! html/parsoid
+
foo
caption
bar

 
-
foo

-
caption

-
bar

 

 !! end
 
 !! test
-3. Bad treebuilder fixup of formatting elt is cleaned up
+3. Treebuilder fixup of formatting elt
 !! options
 parsoid=wt2html,wt2wt
 !! wikitext
 '''foo[[File:Foobar.jpg|thumb|caption]]bar'''
+!! html/php+tidy
+
foo
  
caption
bar
+

 !! html/parsoid
-
foo

-
caption

-
bar

+
foo
caption
bar

 !! end
 
 !! test
-4. Bad treebuilder fixup of formatting elt is cleaned up: formatting tags around captionless images are ignored
+4. Treebuilder fixup of formatting elt: formatting tags around captionless images
 !! options
 parsoid=wt2html,wt2wt
 !! wikitext
 '''[[Image:Foobar.jpg|right|300px]]'''
+!! html/php+tidy
+
Foobar.jpg

 !! html/parsoid
 

 !! end
@@ -27433,16 +27469,17 @@ Encapsulate protected attributes from wt
 ## Currently the p-wrapper is fragile in how it adds / removes transformations.
 ## Having nested or stray pre tags results in the attempt to add duplicates,
 ## causing an assertion fail. This test tries to prevent that situation.
-!!test
+!! test
 Ensure ParagraphWrapper can deal with stray closing pre tags
-!!options
+!! options
 parsoid=wt2html
 !! wikitext
 plain text
 !! html/parsoid
-plain text
-!!end
+
plain text

+!! end
 
+## Remex doesn't account for fostered content.
 !! test
 1. Ensure fostered text content is wrapped in element nodes
 !! options
@@ -27452,9 +27489,10 @@ parsoid=wt2html
 !! html/php+tidy
 hi
ho

 !! html/parsoid
-hi
ho

+
hi
ho

 !! end
 
+## Remex doesn't account for fostered content.
 !! test
 2. Ensure fostered text content is wrapped in element nodes (traps regressions around fostered marker on the element getting lost)
 !! options
@@ -27470,12 +27508,13 @@ parsoid=wt2html,wt2wt
  a
 
 !! html/parsoid
- || ||
-
+
 || ||

+

 
 a
 

 !! end
 
+## Remex doesn't account for fostered content.
 !! test
 Encapsulation properly handles null DSR information from foster box
 !! options
@@ -27483,9 +27522,10 @@ parsoid=wt2html,wt2wt
 !! wikitext
 {{echo|foo
bar
}}
 !! html/parsoid
-foo
bar

+
foo
bar

 !! end
 
+## Remex doesn't account for fostered content.
 !! test
 1. Encapsulate foster-parented transclusion content
 !! options
@@ -27495,7 +27535,7 @@ parsoid=wt2wt,wt2html
 !! html/php+tidy
 foo
bar

 !! html/parsoid
-foo
bar

+
foo
bar

 !! end
 
 !! test
@@ -27553,6 +27593,7 @@ parsoid=wt2wt,wt2html
 
 !! end
 
+## Remex doesn't account for fostered content.
 !! test
 5. Encapsulate foster-parented transclusion content
 !!options
@@ -27562,9 +27603,10 @@ parsoid=wt2wt,wt2html
 !! html/php+tidy
 foo
foo

 !! html/parsoid
-foo
foo

+
foo
foo

 !! end
 
+## Remex doesn't account for fostered content.
 !! test
 6. Encapsulate foster-parented transclusion content
 !! options
@@ -27574,7 +27616,7 @@ parsoid=wt2wt,wt2html
 !! html/php+tidy
 foo
foo
ok

 !! html/parsoid
-foo
foo
ok

+
foo
foo
ok

 !! end
 
 !! test
@@ -27594,6 +27636,7 @@ parsoid=wt2wt,wt2html
 
 !! end
 
+## Remex doesn't account for fostered content.
 # Note that the wt is broken on purpose: the = should be {{=}} if you
 # don't want it to be a template parameter key.
 !! test
@@ -27616,12 +27659,13 @@ parsoid=wt2wt,wt2html
 
 !! html/parsoid
 
a

-{{{1}}}
+
{{{1}}}

 

 b
 

 !! end
 
+## Remex doesn't account for fostered content.
 !! test
 9. Encapsulate foster-parented transclusion content
 !!options
@@ -27632,7 +27676,7 @@ parsoid=wt2wt,wt2html
 hi
hello
 

 !! html/parsoid
-hi
hello

+
hi
hello

 !! end
 
 !! test
@@ -31377,7 +31421,7 @@ __NOTOC__
 
 

-  

   
Table Cell 1Table Cell 2
 foo   

+   foo
 !! end
 
 # Looks like  is not accepted in HTML
-- 
2.20.1