GWicke has uploaded a new change for review.
https://gerrit.wikimedia.org/r/59865
Change subject: Factor out compressHTML method in Util and use it from bugSplit
......................................................................
Factor out compressHTML method in Util and use it from bugSplit
VE output uses innerHTML directly without smart quoting for JSON attributes.
This produces a large diff, which makes it hard to detect actual content
changes.
This patch makes our smart quoting available from Util and uses it in bugSplit
for the editedHtml content.
Change-Id: Id5e628a8af3cacc515b37c7be01ce2f021635e20
---
M js/lib/mediawiki.Util.js
M js/tests/bugSplit.js
2 files changed, 55 insertions(+), 32 deletions(-)
git pull ssh://gerrit.wikimedia.org:29418/mediawiki/extensions/Parsoid
refs/changes/65/59865/1
diff --git a/js/lib/mediawiki.Util.js b/js/lib/mediawiki.Util.js
index f401304..ecaf3b3 100644
--- a/js/lib/mediawiki.Util.js
+++ b/js/lib/mediawiki.Util.js
@@ -1041,6 +1041,45 @@
},
/**
+ * @method compressHTML
+ *
+ * Compress an HTML string by applying some smart quoting
+ *
+ * @param {string} html
+ * @returns {string}
+ */
+compressHTML = function(html) {
+ // now compress our output (and make it more readable) by using
+ // "smart quoting" of attribute values -- using single-quotes
+ // where the contents have a lot of double quotes.
+ // since the output of outerHTML is specified strictly, we know
+ // this regexp is safe. See:
+ //
http://www.whatwg.org/specs/web-apps/current-work/multipage/the-end.html
+ //
http://www.whatwg.org/specs/web-apps/current-work/multipage/syntax.html
+ var smart_quote = function(match, name, equals, value) {
+ if (!equals) { return match; }
+ var decoded = entities.decode(value, 2);
+ // try re-encoding with single-quotes escaped
+ var encoded = decoded.replace(/[&'\u00A0]/g, function(c) {
+ switch(c) {
+ case '&': return '&';
+ case "'": return ''';
+ case '\u00A0': return ' ';
+ }
+ });
+ if (encoded.length >= value.length) { return match; /* no
change */ }
+ return ' '+name+"='"+encoded+"'";
+ };
+ var process_attr_list = function(match, tag, attrs) {
+ attrs = attrs.replace(/ ([^\0-\cZ\s"'>\/=]+)(="([^"]*)")?/g,
+ smart_quote);
+ return tag + attrs + '>';
+ };
+ return html.replace(/(<\w+)((?: [^\0-\cZ\s"'>\/=]+(?:="[^"]*")?)+)>/g,
+ process_attr_list);
+},
+
+/**
* @method serializeNode
*
* Serialize a HTML document.
@@ -1072,35 +1111,7 @@
fictional.appendChild(doc.cloneNode());
html = fictional.innerHTML;
}
- // now compress our output (and make it more readable) by using
- // "smart quoting" of attribute values -- using single-quotes
- // where the contents have a lot of double quotes.
- // since the output of outerHTML is specified strictly, we know
- // this regexp is safe. See:
- //
http://www.whatwg.org/specs/web-apps/current-work/multipage/the-end.html
- //
http://www.whatwg.org/specs/web-apps/current-work/multipage/syntax.html
- var smart_quote = function(match, name, equals, value) {
- if (!equals) { return match; }
- var decoded = entities.decode(value, 2);
- // try re-encoding with single-quotes escaped
- var encoded = decoded.replace(/[&'\u00A0]/g, function(c) {
- switch(c) {
- case '&': return '&';
- case "'": return ''';
- case '\u00A0': return ' ';
- }
- });
- if (encoded.length >= value.length) { return match; /* no
change */ }
- return ' '+name+"='"+encoded+"'";
- };
- var process_attr_list = function(match, tag, attrs) {
- attrs = attrs.replace(/ ([^\0-\cZ\s"'>\/=]+)(="([^"]*)")?/g,
- smart_quote);
- return tag + attrs + '>';
- };
- html = html.replace(/(<\w+)((?: [^\0-\cZ\s"'>\/=]+(?:="[^"]*")?)+)>/g,
- process_attr_list);
- return html;
+ return compressHTML(html);
},
/**
@@ -1115,8 +1126,10 @@
return entities.encode(string, 0 /* xml entities */);
};
+// FIXME gwicke: define this directly
Util.encodeXml = encodeXml;
Util.parseHTML = parseHTML;
+Util.compressHTML = compressHTML;
Util.serializeNode = serializeNode;
Util.normalizeHTML = normalizeHTML;
Util.normalizeOut = normalizeOut;
diff --git a/js/tests/bugSplit.js b/js/tests/bugSplit.js
index f44ab5e..238bb40 100644
--- a/js/tests/bugSplit.js
+++ b/js/tests/bugSplit.js
@@ -2,15 +2,25 @@
* Split up a bug report JSON file into a bunch of files
*/
-var fs = require('fs');
+var fs = require('fs'),
+ Util = require( '../lib/mediawiki.Util.js' ).Util;
function writeFiles ( data ) {
- var keys = Object.keys(data);
+ var keys = Object.keys(data),
+ val;
for ( var i = 0; i < keys.length; i++ ) {
var key = keys[i],
fileName = encodeURIComponent(key);
console.log( 'Creating file ' + fileName );
- fs.writeFileSync(fileName, data[key]);
+
+ val = data[key];
+
+ if (fileName === 'editedHtml') {
+ // apply smart quoting to minimize diff
+ val = Util.compressHTML(val);
+ }
+
+ fs.writeFileSync(fileName, val);
}
}
--
To view, visit https://gerrit.wikimedia.org/r/59865
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: newchange
Gerrit-Change-Id: Id5e628a8af3cacc515b37c7be01ce2f021635e20
Gerrit-PatchSet: 1
Gerrit-Project: mediawiki/extensions/Parsoid
Gerrit-Branch: master
Gerrit-Owner: GWicke <[email protected]>
_______________________________________________
MediaWiki-commits mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
