Issue Bug Resolved

#110782Words with multi-byte characters are cut at 60 bytes when indexing

Benni Mack 09/25/26, 01:11 updated 09/25/26, 16:20

Description

Indexer::analyzeBody() shortens words with byte-based substr() while the header path uses mb_substr(). Words longer than 60 bytes are cut inside a character, and the word hash is built from the broken string, so such words are never found.

Reviews 2

History 26

  1. Benni Mack commented and changed status Under Review Resolved done 0 % 100 % <span data-processor="relativetime" data-value="2026-09-25T16:00:13+00:00" > 09/25/26, 16:00 </span>

    Applied in changeset commit:ad89f371a20859792090dee6c10c51b70c624e11.

  2. Gerrit Code Review pushed patch set 3 for 14.3 to change 96079 status Resolved Under Review <span data-processor="relativetime" data-value="2026-09-25T16:06:56+00:00" > 09/25/26, 16:06 </span>

  3. Benni Mack commented and changed status Under Review Resolved <span data-processor="relativetime" data-value="2026-09-25T16:20:09+00:00" > 09/25/26, 16:20 </span>

    Applied in changeset commit:e3aaf2be9a4546dde7856e7137a6c95f64809d76.