ML-1: archive as not needed

2026-04-24 07:59:49 +01:00
parent 7ad09736cf
commit 38b53592c5
1 changed files with 13 additions and 0 deletions
@@ -0,0 +1,59 @@
+---
+id: ML-1
+title: 'Add default :limit to Maintenance orphan queries'
+status: To Do
+assignee: []
+created_date: '2026-04-20 08:44'
+updated_date: '2026-04-24 06:59'
+labels: []
+dependencies: []
+references:
+  - 'https://github.com/cloud8421/music_library/issues/183'
+priority: low
+---
+
+## Description
+
+<!-- SECTION:DESCRIPTION:BEGIN -->
+_GitHub: created 2026-04-16 · updated 2026-04-16_
+
+## Summary
+
+Two Maintenance queries are unbounded when called without the `:limit` option. Currently invoked only from a Mix task, so not user-facing, but defensively adding a default prevents accidental full-table scans on future call sites.
+
+## Evidence
+
+- `lib/music_library/maintenance.ex:77` — `get_artists_missing_musicbrainz_id/1`
+- `lib/music_library/maintenance.ex:108` — `get_albums_missing_musicbrainz_id/1`
+
+Both accept an `opts` keyword list with `:limit`, but neither applies a default. Callers in `lib/mix/tasks/scrobble/audit.ex` pass explicit limits, so today is fine.
+
+## Fix
+
+Apply a sensible default (e.g. 1000) to the `:limit` option if not provided:
+
+```elixir
+def get_artists_missing_musicbrainz_id(opts \\ []) do
+  limit = Keyword.get(opts, :limit, 1000)
+  # ...
+end
+```
+<!-- SECTION:DESCRIPTION:END -->
+
+## Acceptance Criteria
+<!-- AC:BEGIN -->
+- [ ] #1 Both functions apply a default limit when none is supplied
+- [ ] #2 Existing callers continue to override when they need a different value
+<!-- AC:END -->
+
+## Final Summary
+
+<!-- SECTION:FINAL_SUMMARY:BEGIN -->
+Closed as not-planned (YAGNI).
+
+The proposed fix was to add a default `:limit` (e.g. 1000) to `get_artists_missing_musicbrainz_id/1` and `get_albums_missing_musicbrainz_id/1` in `lib/music_library/maintenance.ex` to defend against future unbounded call sites.
+
+Assessment: the dataset ceiling is <5,000 records and <1,000 artists for the next 2-3 years. An unbounded scan of the `tracks` table grouped by artist/album JSON fields is well within acceptable cost for this size. Adding a default limit would silently cap results on a dataset that should never need capping, and the only current callers (`lib/mix/tasks/scrobble/audit.ex:127,149`) already invoke without `:limit` on purpose.
+
+A follow-up could remove the `:limit` option entirely to simplify the API — not tracked here; revisit if the option starts causing confusion.
+<!-- SECTION:FINAL_SUMMARY:END -->