jenkins-bot has submitted this change and it was merged.
Change subject: Use Wikimedia REST API for accessing page data in Content
Translation
......................................................................
Use Wikimedia REST API for accessing page data in Content Translation
Bug: T92359
Change-Id: I85f5bf4005075326791c87cdadbaeac07316e03c
---
M config.defaults.js
M pageloader/PageLoader.js
2 files changed, 34 insertions(+), 6 deletions(-)
Approvals:
KartikMistry: Looks good to me, approved
jenkins-bot: Verified
diff --git a/config.defaults.js b/config.defaults.js
index bb1b583..c7275b1 100644
--- a/config.defaults.js
+++ b/config.defaults.js
@@ -10,6 +10,7 @@
allowCORS: '*',
// Parsoid API URL
'parsoid.api': 'http://parsoid-lb.eqiad.wikimedia.org',
+ 'restbase.url':
'https://$lang.wikipedia.org/api/rest_v1/page/html/$title',
// Apertium web API URL
'mt.apertium.api': 'http://apertium.wmflabs.org',
'mt.yandex.api': 'https://translate.yandex.net',
diff --git a/pageloader/PageLoader.js b/pageloader/PageLoader.js
index 49a9ac2..2f985ed 100644
--- a/pageloader/PageLoader.js
+++ b/pageloader/PageLoader.js
@@ -13,6 +13,26 @@
conf = require( __dirname + '/../utils/Conf.js' );
/**
+ * Cheap body extraction.
+ *
+ * This is safe as we know that the HTML we are receiving from Parsoid is
+ * serialized as XML.
+ * Restbase does not support body only retrieval of content.
+ * See https://phabricator.wikimedia.org/T95199
+ * @param {string} html
+ * @return {string} body of the html passed, wrapped in <body> tag.
+ */
+function cheapBodyInnerHTML( html ) {
+ var match = /<body[^>]*>([\s\S]*)<\/body>/.exec( html );
+
+ if ( !match ) {
+ throw new Error( 'No HTML body found!' );
+ } else {
+ return '<body>' + match[ 1 ] + '</body>';
+ }
+}
+
+/**
* @class ParsoidPageLoader
*
* @param {string} page
@@ -28,9 +48,14 @@
var url,
deferred = Q.defer();
- url = conf( 'parsoid.api' ) + '/' + this.sourceLanguage + 'wiki/' +
- encodeURIComponent( this.page ) + '?body=1';
-
+ if ( conf( 'restbase.url' ) ) {
+ url = conf( 'restbase.url' )
+ .replace( '$lang', this.sourceLanguage )
+ .replace( '$title', encodeURIComponent( this.page ) );
+ } else {
+ url = conf( 'parsoid.api' ) + '/' + this.sourceLanguage +
'wiki/' +
+ encodeURIComponent( this.page );
+ }
request( url,
function ( error, response, body ) {
if ( error ) {
@@ -41,10 +66,12 @@
deferred.reject( new Error( 'Error while
fetching page: ' + body ) );
return;
}
-
deferred.resolve( {
- body: response.body,
- revision: response.headers[
'content-revision-id' ]
+ body: cheapBodyInnerHTML( response.body ),
+ // Restbase returns revision ID in etag header.
+ // Example:
+ // ETag:
"123456/c4e494da-ee8f-11e4-83a1-8b80de1cde5f"
+ revision: response.headers.etag.split( '/' )[ 0
].replace( '"', '' )
} );
}
);
--
To view, visit https://gerrit.wikimedia.org/r/207039
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: I85f5bf4005075326791c87cdadbaeac07316e03c
Gerrit-PatchSet: 5
Gerrit-Project: mediawiki/services/cxserver
Gerrit-Branch: master
Gerrit-Owner: Santhosh <santhosh.thottin...@gmail.com>
Gerrit-Reviewer: Alexandros Kosiaris <akosia...@wikimedia.org>
Gerrit-Reviewer: KartikMistry <kartik.mis...@gmail.com>
Gerrit-Reviewer: Nikerabbit <niklas.laxst...@gmail.com>
Gerrit-Reviewer: Santhosh <santhosh.thottin...@gmail.com>
Gerrit-Reviewer: jenkins-bot <>
_______________________________________________
MediaWiki-commits mailing list
MediaWiki-commits@lists.wikimedia.org
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
