da12d5d693
Handle ArtistLive.Form saved message in show view
2026-04-19 08:59:04 +01:00
45236e4b10
Convert ErrorIgnorer tests to doctests
2026-04-19 07:50:39 +01:00
fb1b60373d
Fix max width issue for release selector on mobile
2026-04-18 10:57:14 +01:00
f6bcf3d970
Partial fix for barcode scan camera persisting after disconnection
2026-04-18 10:50:41 +01:00
b37162fce3
Fix colors of mixed badges
2026-04-17 22:30:27 +01:00
94143010c0
Fix doubly-wrapped safe tuple in artist bio render
2026-04-17 16:00:22 +01:00
e6a8488137
Use Notes.change_note/2 instead of Note.changeset/2 in notes component
...
Closes #171
2026-04-17 12:56:08 +01:00
bbe1722c9f
Handle unencodable structs in QueryReporter params
2026-04-17 07:10:45 +01:00
fbc548e903
Explicitly trust Wikipedia's HTML
...
Closes #168
2026-04-16 23:14:00 +01:00
464e627a63
Remove unnecessary padding from detail views
2026-04-16 22:54:39 +01:00
6b35b5cde5
Trigger search on Enter in search inputs inside forms
2026-04-16 15:15:47 +01:00
58c60193f2
Use pipe
2026-04-16 10:38:08 +01:00
840547c509
Speed up structured modal animation
2026-04-16 09:31:30 +01:00
fea1231304
Replace refresh scrobble list icon
...
Makes it for a better animation
2026-04-15 22:52:28 +01:00
7899f621f3
Preserve rule picker state across parent re-renders
2026-04-15 15:11:17 +01:00
2124c982f7
Handle possible non-existent record id in search filter
2026-04-15 15:10:28 +01:00
c816f2de4a
Add scrobbled tracks search parser
...
Introduces ListeningStats.SearchParser with NimbleParsec supporting
record:, album_mbid:, artist_mbid:, artist:, album:, track: filters.
Links scrobble count on record page to filtered scrobbled tracks view.
2026-04-15 14:56:38 +01:00
851cb36858
Fix dl accessibility violations on stats dashboard
2026-04-15 14:16:24 +01:00
4194fb8b6d
Update dependencies
...
oban_web 2.12.2 => 2.12.3
2026-04-15 13:50:52 +01:00
405235e2b3
Add picker to create scrobble rules on the fly
2026-04-15 10:53:58 +01:00
b9612a2a26
Fix records on this day for partial release dates
...
SQLite interprets bare years (e.g. "1970") as Julian day numbers
in strftime, producing wrong month-day matches. Filter to only
full YYYY-MM-DD dates before comparing.
2026-04-15 07:29:58 +01:00
ebd9e5445f
Fix how long ago for records with year release dates
2026-04-15 07:20:01 +01:00
d75508531e
Generate parse_type/parse_format from module attributes
2026-04-14 16:56:04 +01:00
2bb755b1c8
Parse matching record fields in Records.Record
2026-04-14 16:51:06 +01:00
8db324d74a
Extract shared badge/dropdown to ScrobbleComponents
2026-04-14 16:50:40 +01:00
888b5a68e6
Update all callers to use matching_records
...
Migrate StatsLive.Index, ScrobbledTracksLive.Index, and TopAlbums to
consume matching_records list instead of scalar record IDs. Remove the
derive_legacy_record_ids bridge function. TopAlbums now uses a dropdown
for multi-record badges with play count display.
2026-04-14 16:14:59 +01:00
9ef918a621
Update record_status_badges with matching_records and dropdown
2026-04-14 16:10:57 +01:00
2c6e6e7ca2
Return matching_records list from top_albums_attach_metadata
...
Apply the same release-group JSON aggregation pattern to top albums
queries. Add JSON parsing in get_top_albums/1 and
get_top_albums_by_days/2.
2026-04-14 16:05:51 +01:00
cfd267627c
Return matching_records list from tracks_with_record_info_query
...
Replace scalar collected_record_id/wishlisted_record_id subqueries with
a single matching_records JSON array grouped by release group
(records.musicbrainz_id). Add parse_matching_records/1 for JSON parsing
and derive_legacy_record_ids/2 as a temporary bridge for existing
templates. Update tests to use shared musicbrainz_id fixtures.
2026-04-14 14:17:02 +01:00
9f4135661e
Replace scalar record IDs with matching_records JSON array in tracks query
...
Replaces collected_record_id and wishlisted_record_id correlated
subqueries in tracks_with_record_info_query with a single
matching_records json_group_array subquery that returns all records
sharing the same release ID. Adds parse_matching_records/1 for JSON
deserialization. Updates recent_activity/2 and list_tracks/1 to parse
the JSON at the boundary. Derives legacy collected_record_id and
wishlisted_record_id from matching_records so existing LiveViews keep
working. The get_top_albums test is temporarily skipped pending the
top_albums_attach_metadata update.
2026-04-14 14:12:14 +01:00
e43bccb9bc
Extract shared index logic into LiveHelpers.IndexActions
...
Closes #150
2026-04-13 23:09:22 +01:00
4fa13d5422
Reset chat state when navigating between entities
...
The Chat component kept stale messages and chat references
when patching to a different artist/record because update/2
only refreshed the chat list, not the active chat state.
2026-04-13 18:32:19 +01:00
6a6410bcaa
Have the LLM use square bracketed links for internal references
2026-04-13 14:24:36 +01:00
96bb0fc403
Try and control ChatGPT's tone and "helpfulness"
2026-04-13 14:09:43 +01:00
f7628267b3
Add @moduledoc to OpenAI.API
2026-04-13 13:15:43 +01:00
bdc1cd3992
Prevent callback return value from halting stream
...
The delta branch in decode_responses_event now explicitly returns
:ok after invoking the callback, so a callback returning {:error, _}
cannot be mistaken for a stream-level error by decode_events.
2026-04-13 13:14:33 +01:00
6fa640f0f6
Remove unreachable catch clause in do_chat_stream
2026-04-13 13:12:58 +01:00
9b44b7050f
Count unique catalog entries in collection summary
...
The record_count now reflects the number of grouped catalog lines
the model sees, not raw DB rows. Two copies of the same album on
different formats count as 1 catalog entry.
2026-04-13 13:12:30 +01:00
76438360d7
Load collection summary asynchronously on mount
2026-04-13 13:11:15 +01:00
5512ac0837
Hide button text on mobile
2026-04-13 12:53:57 +01:00
54dab5fbdf
Use gpt 5.1 to cater for collection summary context size
2026-04-13 12:43:42 +01:00
f9448cac6f
Revert "Use smaller model so that we don't exhaust tokens"
...
This reverts commit 9c0e11bc84
2026-04-13 12:34:17 +01:00
8108645718
Shrink collection summary to fit OpenAI token limit
...
Deduplicate records by musicbrainz_id (merging formats)
and cap genres to 3 per record, reducing the summary from
~32K to ~24K estimated tokens.
2026-04-13 12:33:15 +01:00
9c0e11bc84
Use smaller model so that we don't exhaust tokens
2026-04-13 12:06:24 +01:00
f73be3f881
Handle OpenAI api streaming error messages
2026-04-13 11:54:05 +01:00
3b8415d8c4
Add Collection Chat to universal search
2026-04-13 10:15:43 +01:00
a509248625
Wire up Chat component in CollectionLive.Index
2026-04-13 10:12:49 +01:00
34f484cc81
Add CollectionChat StreamProvider
2026-04-13 10:09:50 +01:00
24a3cc9265
Add Collection.collection_summary/0
2026-04-13 10:08:01 +01:00
d845ab93dd
Add collection sentinel UUID to Chats context
2026-04-13 10:06:16 +01:00
Claudio Ortolina