EBernhardson has uploaded a new change for review.
https://gerrit.wikimedia.org/r/286238
Change subject: Handle urls using title= query string
......................................................................
Handle urls using title= query string
Change-Id: I96011f355c3cbd1162e00e2f10c7ed2e86ba9b93
---
M src/RelevanceScoring/Import/ImportedResult.php
M tests/unit/RelevanceScoring/Import/ImportedResultTest.php
2 files changed, 25 insertions(+), 6 deletions(-)
git pull ssh://gerrit.wikimedia.org:29418/wikimedia/discovery/discernatron
refs/changes/38/286238/1
diff --git a/src/RelevanceScoring/Import/ImportedResult.php
b/src/RelevanceScoring/Import/ImportedResult.php
index 9603a48..c531fd4 100644
--- a/src/RelevanceScoring/Import/ImportedResult.php
+++ b/src/RelevanceScoring/Import/ImportedResult.php
@@ -35,13 +35,24 @@
// make the bold assumption wikimedia wikis all
// prefix with /wiki/
$prefix = '/wiki/';
- if ($prefix !== substr($path, 0, strlen($prefix))) {
- throw new \Exception("Invalid url: $url");
+ if ($prefix === substr($path, 0, strlen($prefix))) {
+ $titlePart = substr($path, strlen($prefix));
+ $title = urldecode(strtr($titlePart, '_', ' '));
+ if (!empty($title)) {
+ return new self($source, $title, $snippet, $position);
+ }
}
- $titlePart = substr($path, strlen($prefix));
- $title = urldecode(strtr($titlePart, '_', ' '));
- return new self($source, $title, $snippet, $position);
+ $query = parse_url($url, PHP_URL_QUERY);
+ if ($query) {
+ parse_str($query, $decoded);
+ if (!empty($decoded['title'])) {
+ $title = strtr($decoded['title'], '_', ' ');
+ return new self($source, $title, $snippet, $position);
+ }
+ }
+
+ throw new \Exception("Invalid url: $url");
}
public function getSource()
diff --git a/tests/unit/RelevanceScoring/Import/ImportedResultTest.php
b/tests/unit/RelevanceScoring/Import/ImportedResultTest.php
index 506096c..bea8eef 100644
--- a/tests/unit/RelevanceScoring/Import/ImportedResultTest.php
+++ b/tests/unit/RelevanceScoring/Import/ImportedResultTest.php
@@ -18,6 +18,14 @@
'Fuller\'s Brewery',
'https://en.wikipedia.org/wiki/Fuller%27s_Brewery',
),
+ 'oddly formed url with query string' => array(
+ 'Talk:SL-1',
+ 'https://en.wikipedia.org/wiki?title=Talk:SL-1',
+ ),
+ 'query string with encoded parts' => array(
+ 'Foo & Bar',
+
'https://en.wikipedia.org/w/index.php?title=Foo_%26_Bar',
+ ),
);
}
@@ -26,7 +34,7 @@
*/
public function testCreateFromURL($title, $url)
{
- $result = ImportedResult::createFromURL('unitTest', $url, 1);
+ $result = ImportedResult::createFromURL('unitTest', $url, '', 1);
$this->assertEquals('unitTest', $result->getSource());
$this->assertEquals($title, $result->getTitle());
--
To view, visit https://gerrit.wikimedia.org/r/286238
To unsubscribe, visit https://gerrit.wikimedia.org/r/settings
Gerrit-MessageType: newchange
Gerrit-Change-Id: I96011f355c3cbd1162e00e2f10c7ed2e86ba9b93
Gerrit-PatchSet: 1
Gerrit-Project: wikimedia/discovery/discernatron
Gerrit-Branch: master
Gerrit-Owner: EBernhardson <[email protected]>
_______________________________________________
MediaWiki-commits mailing list
[email protected]
https://lists.wikimedia.org/mailman/listinfo/mediawiki-commits
