jenkins-bot has submitted this change and it was merged. (
https://gerrit.wikimedia.org/r/404510 )
Change subject: summary: fix mainpage handling
......................................................................
summary: fix mainpage handling
The issue was that several non-English main pages are not in the
main name space. In those cases we were returning a 204 HTTP code and
no content.
Change-Id: Ie15487184a7f9fc08603fc42cfad3aeac6642dcc
---
M lib/summary.js
M test/features/summary/pagecontent.js
2 files changed, 37 insertions(+), 7 deletions(-)
Approvals:
jenkins-bot: Verified
Mholloway: Looks good to me, approved
diff --git a/lib/summary.js b/lib/summary.js
index 24a9e4a..62c57ce 100644
--- a/lib/summary.js
+++ b/lib/summary.js
@@ -7,6 +7,7 @@
const NS_MAIN = 0;
const SUMMARY_NS_WHITELIST = [ NS_MAIN ];
+const EMPTY_EXTRACTS = { extract: '', extract_html: '' };
/**
* Builds a dictionary containing the various forms of a page title that a
client may need.
@@ -64,7 +65,7 @@
}
/**
- * Get the page summary type.
+ * Gets the page summary type.
* @param {!Object} meta page metadata from MW API
* return {!String} the summary type (one of 'standard', 'disambiguation', or
'mainpage')
*/
@@ -80,6 +81,26 @@
return 'standard';
}
+/**
+ * Builds the extract values.
+ * @param {!Document} doc a DOM Document with the page content
+ * @param {!boolean} isMainPage if the page is a main page
+ * return {!Object {extract, extract_html} } the extract values
+ */
+function buildExtracts(doc, isMainPage) {
+ if (isMainPage) {
+ return EMPTY_EXTRACTS;
+ } else {
+ const leadSectionDoc = parsoidSections.justLeadSection(doc);
+ const intro = transforms.extractLeadIntroduction(leadSectionDoc);
+ if (intro.length) {
+ return transforms.summarize(intro);
+ } else {
+ return EMPTY_EXTRACTS;
+ }
+ }
+}
+
/*
* Build a page summary
* @param {!String} domain the request domain
@@ -90,18 +111,16 @@
* @return {!Object} a summary 2.0 spec-compliant page summary object
*/
function buildSummary(domain, title, html, revTid, meta) {
- if (shouldReturn204(meta)) {
+ if (!meta.mainpage && shouldReturn204(meta)) {
return { code: 204 };
}
const doc = domino.createDocument(html);
- const leadSectionDoc = parsoidSections.justLeadSection(doc);
- const intro = transforms.extractLeadIntroduction(leadSectionDoc);
- const summary = intro.length ? transforms.summarize(intro) : { extract:
'', extract_html: '' };
+ const extracts = buildExtracts(doc, meta.mainpage);
return Object.assign({
code: 200,
- type : getSummaryType(meta),
+ type: getSummaryType(meta),
title: meta.normalizedtitle,
displaytitle: meta.displaytitle,
namespace: { id: meta.ns, text: meta.nsText },
@@ -121,7 +140,7 @@
},
content_urls: buildContentUrls(domain, title, meta),
api_urls: buildApiUrls(domain, title, meta),
- }, summary);
+ }, extracts);
}
module.exports = {
diff --git a/test/features/summary/pagecontent.js
b/test/features/summary/pagecontent.js
index 3ffa217..35933b8 100644
--- a/test/features/summary/pagecontent.js
+++ b/test/features/summary/pagecontent.js
@@ -57,6 +57,17 @@
});
});
+ it('main page in non-mainspace should also return type: \'mainpage\'', ()
=> {
+ const uri = localUri('Wikipedia:Hauptseite', 'de.wikipedia.org');
+ return preq.get({ uri })
+ .then((res) => {
+ assert.deepEqual(res.status, 200);
+ assert.deepEqual(res.body.type, 'mainpage', 'type should be
\'mainpage\'');
+ assert.deepEqual(res.body.extract, '', 'should send empty
plaintext extract');
+ assert.deepEqual(res.body.extract_html, '', 'should send empty
html extract');
+ });
+ });
+
function should204(uri) {
return preq.get({ uri })
.then((res) => {
--
To view, visit https://gerrit.wikimedia.org/r/404510
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: merged
Gerrit-Change-Id: Ie15487184a7f9fc08603fc42cfad3aeac6642dcc
Gerrit-PatchSet: 1
Gerrit-Project: mediawiki/services/mobileapps
Gerrit-Branch: master
Gerrit-Owner: BearND <[email protected]>
Gerrit-Reviewer: Fjalapeno <[email protected]>
Gerrit-Reviewer: Jdlrobson <[email protected]>
Gerrit-Reviewer: Mholloway <[email protected]>
Gerrit-Reviewer: Mhurd <[email protected]>
Gerrit-Reviewer: Ppchelko <[email protected]>
Gerrit-Reviewer: jenkins-bot <>
_______________________________________________
MediaWiki-commits mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
