jenkins-bot has submitted this change and it was merged. (
https://gerrit.wikimedia.org/r/379902 )
Change subject: Remove data-mw attributes before parsing summary
......................................................................
Remove data-mw attributes before parsing summary
Bug: T176521
Change-Id: Ifd0c47114973fd7c1cc1c2231a52947b5dfcbe34
---
M lib/transformations/summarize.js
M test/lib/transformations/summarize.js
2 files changed, 7 insertions(+), 0 deletions(-)
Approvals:
BearND: Looks good to me, approved
jenkins-bot: Verified
diff --git a/lib/transformations/summarize.js b/lib/transformations/summarize.js
index 342371c..bf9e205 100644
--- a/lib/transformations/summarize.js
+++ b/lib/transformations/summarize.js
@@ -3,6 +3,7 @@
const domino = require('domino');
const flattenElements = require('./flattenElements');
const rmElementsWithSelector = require('./rmElementsWithSelector');
+const removeAttributes = require('./removeAttributes');
/**
* Recursively discard any parentheticals that themselves are inside
parentheticals
@@ -29,6 +30,7 @@
module.exports = function(html) {
const doc = domino.createDocument(html);
flattenElements(doc, 'a');
+ removeAttributes(doc, '*', ['data-mw']);
rmElementsWithSelector(doc, '.mw-ref, .reference');
rmElementsWithSelector(doc, '.noexcerpts');
rmElementsWithSelector(doc, 'math');
diff --git a/test/lib/transformations/summarize.js
b/test/lib/transformations/summarize.js
index 75d3532..b58e3a3 100644
--- a/test/lib/transformations/summarize.js
+++ b/test/lib/transformations/summarize.js
@@ -43,6 +43,11 @@
'<p>The Planck–Einstein relation connects the particulate
photon energy <span class=\"texhtml \"><i>E</i></span> with its associated wave
frequency <span class=\"texhtml \"><i>f</i></span>:</p>\n\n<dl id=\"mwmQ\"><dd
id=\"mwmg\"><span class=\"mwe-math-element\"><span
class=\"mwe-math-mathml-inline mwe-math-mathml-a11y\" style=\"display:
none;\"><math xmlns=\"http://www.w3.org/1998/Math/MathML\";>\n <semantics>\n
<mrow class=\"MJX-TeXAtom-ORD\">\n <mstyle displaystyle=\"true\"
scriptlevel=\"0\">\n <mi>E</mi>\n <mo>=</mo>\n
<mi>h</mi>\n <mi>f</mi>\n </mstyle>\n </mrow>\n <annotation
encoding=\"application/x-tex\">{\\displaystyle E=hf}</annotation>\n
</semantics>\n</math></span><img
src=\"https://wikimedia.org/api/rest_v1/media/math/render/svg/f39fac3593bb1e2dec0282c112c4dff7a99007f6\";
class=\"mwe-math-fallback-image-inline\" aria-hidden=\"true\"
style=\"vertical-align: -0.671ex; width:7.533ex;
height:2.509ex;\"></span></dd></dl>',
'<p>The Planck–Einstein relation connects the particulate
photon energy <span class=\"texhtml \"><i>E</i></span> with its associated wave
frequency <span class=\"texhtml \"><i>f</i></span>:</p>\n\n<dl id=\"mwmQ\"><dd
id=\"mwmg\"><span class=\"mwe-math-element\"><img
src=\"https://wikimedia.org/api/rest_v1/media/math/render/svg/f39fac3593bb1e2dec0282c112c4dff7a99007f6\";
class=\"mwe-math-fallback-image-inline\" aria-hidden=\"true\"
style=\"vertical-align: -0.671ex; width:7.533ex;
height:2.509ex;\"></span></dd></dl>'
],
+ // Any parentheticals inside a data-mw attribute are ignored.
+ [
+ '<p><b>Shakira Isabel Mebarak Ripoll</b> (<small data-mw="(t)
<!--Arabic (Spanish pronunciation)">pronounced<span> </span></small><span
class="IPA"><span>[(t)ʃaˈkiɾa isaˈβel meβaˈɾak riˈpol]</span></span>;
<span><span class="ipa_button"></span><span class="nowrap
mcs-ipa"><small>English: </small><span class="IPA nopopups
noexcerpt"><span>/<span style="border-bottom:1px
dotted"><span>ʃ</span><span>ə</span><span>ˈ</span><span>k</span><span>iː</span><span>r</span><span>ə</span></span>/</span></span></span></span>;
born 2 February 1977) is a Colombian singer, songwriter, dancer.</p>',
+ '<p><b>Shakira Isabel Mebarak Ripoll</b> is a Colombian
singer, songwriter, dancer.</p>'
+ ],
// Any content in parentheticals is stripped and no double spaces
are left in the output
[
'<p><b>Epistemology</b> (<span class="nowrap"><span class="IPA
nopopups noexcerpt"><span>/<span style="border-bottom:1px dotted"><span
title="/ɪ/ or /ə/ \'e\' in \'roses\'">ᵻ</span><span title="/ˌ/ secondary stress
follows">ˌ</span><span title="\'p\' in \'pie\'">p</span><span title="/ɪ/ short
\'i\' in \'bid\'">ɪ</span><span title="\'s\' in \'sigh\'">s</span><span
title="\'t\' in \'tie\'">t</span><span title="/ɪ/ or /ə/ \'e\' in
\'roses\'">ᵻ</span><span title="/ˈ/ primary stress follows">ˈ</span><span
title="\'m\' in \'my\'">m</span><span title="/ɒ/ short \'o\' in
\'body\'">ɒ</span><span title="\'l\' in \'lie\'">l</span><span title="/ə/ \'a\'
in \'about\'">ə</span><span title="/dʒ/ \'j\' in \'jam\'">dʒ</span><span
title="/i/ \'y\' in \'happy\'">i</span></span>/</span></span><small
class="nowrap metadata"> (<span class="unicode haudio"><span
class="fn"><span style="white-space:nowrap"><span><img alt="About this sound"
src="//upload.wikimedia.org/wikipedia/commons/thumb/8/8a/Loudspeaker.svg/11px-Loudspeaker.svg.png"
width="11" height="11"
srcset="//upload.wikimedia.org/wikipedia/commons/thumb/8/8a/Loudspeaker.svg/17px-Loudspeaker.svg.png
1.5x,
//upload.wikimedia.org/wikipedia/commons/thumb/8/8a/Loudspeaker.svg/22px-Loudspeaker.svg.png
2x" data-file-width="20"
data-file-height="20"></span> </span><span>listen</span></span></span>)</small></span>;
from <span>Greek</span> <span lang="gre" xml:lang="gre"><span
>ἐπιστήμη</span><i>, epistēmē</i></span>, meaning \'knowledge\', and <span
lang="" xml:lang=""><span>λόγος</span><i>, <span>logos</span></i></span>,
meaning \'logical discourse\') is the <span>branch</span> of
<span>philosophy</span> concerned with the theory of
<span>knowledge</span>.</p>',
--
To view, visit https://gerrit.wikimedia.org/r/379902
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: Ifd0c47114973fd7c1cc1c2231a52947b5dfcbe34
Gerrit-PatchSet: 4
Gerrit-Project: mediawiki/services/mobileapps
Gerrit-Branch: master
Gerrit-Owner: Jdlrobson <jrob...@wikimedia.org>
Gerrit-Reviewer: BearND <bsitzm...@wikimedia.org>
Gerrit-Reviewer: Dbrant <dbr...@wikimedia.org>
Gerrit-Reviewer: Fjalapeno <cfl...@wikimedia.org>
Gerrit-Reviewer: GWicke <gwi...@wikimedia.org>
Gerrit-Reviewer: Jdlrobson <jrob...@wikimedia.org>
Gerrit-Reviewer: Mholloway <mhollo...@wikimedia.org>
Gerrit-Reviewer: Mhurd <mh...@wikimedia.org>
Gerrit-Reviewer: Mobrovac <mobro...@wikimedia.org>
Gerrit-Reviewer: Ppchelko <ppche...@wikimedia.org>
Gerrit-Reviewer: jenkins-bot <>
_______________________________________________
MediaWiki-commits mailing list
MediaWiki-commits@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
