@recaptime-dev's working patches + fork for Phorge, a community fork of Phabricator. (Upstream dev and stable branches are at upstream/main and upstream/stable respectively.)
hq.recaptime.dev/wiki/Phorge
phorge
phabricator
Identify compound short search tokens in the form "xx.yy" as unqueryable in the search UI
Summary:
Ref T12928. The index doesn't work for these, so show the user that there's a problem and drop the terms.
This doesn't fix the problem, but makes the behavior more clear.
Test Plan:
{F5053703}
{F5053704}
Reviewers: chad
Reviewed By: chad
Maniphest Tasks: T12928
Differential Revision: https://secure.phabricator.com/D18254
+19
-1
+19
-1
src/applications/search/fulltextstorage/PhabricatorMySQLFulltextStorageEngine.php
+19
-1
src/applications/search/fulltextstorage/PhabricatorMySQLFulltextStorageEngine.php
···
235
235
$value = $stemmer->stemToken($value);
236
236
}
237
237
238
-
if (phutil_utf8_strlen($value) < $min_length) {
238
+
if ($this->isShortToken($value, $min_length)) {
239
239
$fulltext_token->setIsShort(true);
240
240
continue;
241
241
}
···
547
547
$stopwords = array_fuse($stopwords);
548
548
549
549
return array($min_len, $stopwords);
550
+
}
551
+
552
+
private function isShortToken($value, $min_length) {
553
+
// NOTE: The engine tokenizes internally on periods, so terms in the form
554
+
// "ab.cd", where short substrings are separated by periods, do not produce
555
+
// any queryable tokens. These terms are meaningful if at least one
556
+
// substring is longer than the minimum length, like "example.py". See
557
+
// T12928.
558
+
559
+
$parts = preg_split('/[.]+/', $value);
560
+
561
+
foreach ($parts as $part) {
562
+
if (phutil_utf8_strlen($part) >= $min_length) {
563
+
return false;
564
+
}
565
+
}
566
+
567
+
return true;
550
568
}
551
569
552
570
}