{"version":"2.1.232","anchor":"transcript-compaction-planner-split-out-of-the-writer","canonical_anchor":"transcript-compaction-planner-split-out-of-the-writer","heading":"Transcript compaction planner split out of the writer","tier":"internal","area":"Elsewhere","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.232\/e\/transcript-compaction-planner-split-out-of-the-writer","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.232","markdown":"### Transcript compaction planner split out of the writer\n\nThe logic deciding which old transcript messages to drop was split into scan, plan and filter steps.\n\n**What**\n\nThe code that decides which on-disk transcript messages to drop was extracted into separate scan, plan and filter functions, with the same abort telemetry as before when a preserved message id is missing or the preserved-segment walk breaks.\n\n**Evidence**\n\n`preserved_walk_broken`\n\n- Area: Elsewhere\n- Tier: Under the hood\n- Useful: 1\/5\n- Signal: 1\/5"}