Changes in MemoryIndexTest.php

bc12b8fead60f0630a4d5ef3de17d2c7722aa5cf - fix(search): read split index suffixes correctly

Sun, 05 Jul 2026 08:25:22 +0000

fix(search): read split index suffixes correctlyAbstractIndex::max() used /(\d)+\.idx$/ to determine the highest splitindex suffix. For filenames like w10.idx or w12.idx that regex capturedonly the last digit, so the maximum token-length group was truncated to 9.Wildcard searches therefore skipped all fulltext index shards for wordswith 10 or more characters.Tighten the match to the basename and require the current index namefollowed immediately by digits: ^(\d+)\.idx$. This captures the fullnumeric suffix and also avoids counting unrelated index families thatshare the same prefix, such as treating wiki2.idx as a numbered shard ofthe w index.Add regressions for both behaviors: multi-digit suffixes are parsedcorrectly, and same-prefix indexes are ignored when determining themaximum shard number. List of files: /dokuwiki/_test/tests/Search/Index/MemoryIndexTest.php

06053dca2fac9a1da4eb1accf8c2488942da5d2a - SearchIndex: remove write side effect from retrieveRow()

Fri, 10 Apr 2026 19:06:30 +0000

SearchIndex: remove write side effect from retrieveRow()retrieveRow() padded the index file when the requested RID was beyondthe current length. This was an optimization for subsequent changeRow()calls, but changeRow() already handles padding on its own. The sideeffect was also inconsistent with retrieveRows() which is a pure read.Added a cross-index integration test verifying RID consistency acrossentity, token, frequency and reverse indexes when multiple entitiesshare tokens. List of files: /dokuwiki/_test/tests/Search/Index/MemoryIndexTest.php

ede4646658cf51245060332d97a319a39c788ea1 - SearchIndex: reorganize and expand test suite

Mon, 06 Apr 2026 16:45:11 +0000

SearchIndex: reorganize and expand test suiteMove all Search tests from _test/tests/inc/Search/ to _test/tests/Search/to match the dokuwiki\test autoloader convention. Fix namespaces fromtests\* to dokuwiki\test\* so all tests work in isolation.Extract inline test helpers into separate autoloadable mock files:TestDirectCollection → MockDirectCollection, TestLookupCollection →MockLookupCollection, TestFrequencyCollection → MockFrequencyCollection.Rename AbstractIndexTest → AbstractIndexTestCase to fix PHPUnit warningabout abstract classes with Test suffix.Replace dead xxxRealWord() with proper testWildcardSearch() verifyingexact token matches and frequencies for all three wildcard types.Add testTokenizedPageSearch() using a dedicated test data file. AddtestNoMatchReturnsEmptyFrequencies() which exposed a bug in Term whereuninitialized $tokens/$frequencies caused crashes on zero-match terms.Replace fulltext_query.test.php with modern QueryParserTest in theSearch\Query namespace.Add new test files:- LockTest: acquire/release, reference counting, stale lock override, foreign lock rejection, releaseAll, independent locks- NamespacePredicateTest: filter/exclude, sub-namespaces, partial prefix safety, empty sets, score preservation- PageSetTest: intersect, unite, subtract, isEmpty- QueryEvaluatorTest: word lookups, AND/OR/NOT, namespace filtering, combined queries, partial namespace prefix safetyFix Term.php: initialize $tokens and $frequencies to [] instead of null. List of files: /dokuwiki/_test/tests/Search/Index/MemoryIndexTest.php