- Every question asked of Claude is recorded: when, the kind of list, the model, the tokens read and written (also for an answer that was no name but was billed) and how it ended: a name made, a name taken from the kept ones, or a failure and why. Nothing of the facts that were sent is recorded. Records older than a year are removed (migration 0022).
- GET /providers/anthropic/usage (admin) gives the totals of the last days (1 to 365, 30 by default), a row for each day with calls and an estimate of the cost in dollars, worked out from the tokens and the price of the model. It says it is an estimate and flags calls to a model whose price is not known.
- A Costs section in the admin area shows the last 7, 30 or 90 days.
- Keeping a mix, a smart playlist or a suggested smart playlist as a playlist with auto_name gives it the name and the line of description that Claude makes from its songs. The name sent stays the fallback when the provider is off, or Claude fails or declines, so a failure never stops a playlist from being made.
- A smart playlist can be made with no name when Claude is on: it names it from the songs the rules give. Without Claude a name is still needed.
- GET /naming says whether names are automatic, so the editor does not ask for a name that Claude will make.
- The web app always asks for an automatic name and sends the language of the page.
- A smart playlist rule, plays_within_days, that counts only the plays of the last N days where plays are counted: the minimum of plays and the most and least played orders. The other listening rules are not changed.
- Two more suggested smart playlists on it: on-rotation (the most played of the last 30 days, at least twice) and top-year (the same over 365 days). most-played stays as the plays of all time.
- The editor has the new rule and the page says it in words. Texts in English and Italian.
- Where a playlist is renamed, and where a smart playlist is edited, a button asks the server for a name in the language of the page. It fills the name, fills the description when it is empty, and says if the name came from Claude or from this server. A failed request says why and changes nothing.
- The Claude card in the providers page has a key, a test and the switch, but no contact and no data to take back.
- A pipeline over the facts about a list of songs (up to 5 genres, up to 8 artist names, how many songs, a few words about the rules of a smart playlist, the language). A local namer builds a name from the genres and the artists on this server and is always there; Claude is asked first only when an admin turned the new anthropic provider on with an API key. A failure or a refusal gives the local name. What Claude answers is kept under a hash of the facts, so the same question is not asked or paid twice.
- The anthropic provider: a write-only key (kept as typed, it is case-sensitive), a test that reads one model and sends no text, and a plain statement of what is sent. It uses the official Go SDK with nothing taken from the environment of the server (no key, profile, base URL or extra headers), no redirects, no system headers, and an answer that is cleaned and bounded.
- POST /playlists/{id}/suggest-name, POST /mixes/{id}/suggest-name and POST /smart-playlists/{id}/suggest-name answer with a name, a description and where it came from.
- A button in the sidebar folds it to a narrow bar with the icon of each page, and opens it again. The choice is remembered in the browser, which may refuse to keep it.
- Every link keeps its name for screen readers and gets a tooltip when only its icon shows.
- On a phone, where the sidebar is a row of links, nothing changes.
- A fixed set that is there with no setup: recently added, never played, most played, not heard for a while, high quality files and long tracks. The rules are the same for everyone and are worked out on each one's own library and history, like the mixes; nothing is stored, so they follow the library.
- GET /smart-playlists/suggested, GET /smart-playlists/suggested/{key}/tracks and POST /smart-playlists/suggested/{key}/playlist.
- The Smart playlists page begins with them, with names and descriptions in the language of the page. A suggested one can be played, shuffled again, kept as a playlist, or made one's own: a copy of its rules opens in the editor.
- The server says which albums stand for a mix (album_ids): up to 4 albums that have a cover, by different artists, with most of the songs the mix draws from. The list gives them for every mix and a mix gives its own.
- The Mixes page shows the collage on each card (a 2 by 2 collage with four covers, the first one alone with fewer, the rings with none), and the page of a mix shows it in its header.
- Nothing is fetched for it: a mix has a picture once the covers of its albums were looked up.
- A Smart playlists page with the ones I made and a way to make a new one.
- A page for a smart playlist: its rules in words, its songs, Play, Shuffle again (for the random order), Save as playlist (keeps what is on screen), Edit and Delete.
- An editor for a new one and for one that exists, with a live preview of the songs the rules give: genres (chips, with the library's genres offered), file format and bitrate, when it was added, listening, taste, length, order and limit. What the editor has no field for (artists, albums, skipped lately, another format) is kept when a playlist is changed.
- Texts in English and Italian, and tests for the list, the page, the editor and the preview.
Each genre with the number of playable songs, the biggest first and by name among equals. The spellings the files have are one genre, written the way the rules compare them, and tags such as other or unknown are left out.
A smart playlist is a set of rules, an order and a limit, kept under a name and worked out again every time it is opened.
- Rules (all that are set must hold): genres, artists and albums (is, or is not), file formats and a minimum bitrate of the same file, added in the last N days, never played, played or not played in the last N days, played at least N times, skipped or not skipped lately, only liked artists or without the avoided ones, and the length of the song.
- Orders: random (the same all day, a variant gives another), most played, least played, recently added, by title. A limit of 50 by default, 500 at most.
- GET/POST /smart-playlists, GET/PATCH/DELETE /smart-playlists/{id}, GET /smart-playlists/{id}/tracks, POST /smart-playlists/preview to try rules without keeping them, and POST /smart-playlists/{id}/playlist to keep what is shown as a normal playlist.
- Only the rules are stored (migration 0020); the songs come from the files and from the owner's own history and taste. Nothing is sent to a provider, and a smart playlist is its owner's alone. A user can have 200.
- Mixes and smart playlists share the way a list of songs becomes a playlist.
When MusicBrainz merges two entries, the old id answers with a redirect to the new one. The client refused every redirect, so the description of that artist failed at every run of the job.
The client for MusicBrainz now follows a redirect only to https://musicbrainz.org/ws/2/artist/<id> or /release-group/<id>, with no credentials or port. Any other redirect is still refused. The new id is used for that request only.
- A Mixes link in the sidebar and a page with one card for each mix: the ones made from listening first, then one for each genre.
- A page for a mix: its songs, Play, Shuffle again (the next variant, one at a time) and Save as playlist (keeps the variant on screen, named with the title, with a link to the new playlist).
- Mix names are translated from the kind and the genre, since the server title is always English.
- Texts in English and Italian; tests for the list, the page, shuffle, save and the errors.
GET /mixes lists the mixes a listener has, GET /mixes/{id}?variant= gives one with its
songs, and POST /mixes/{id}/playlist keeps one as a playlist of their own.
- Genre mixes: one for each genre with at least 30 songs by 4 artists, the genres the
listener plays most first. Spellings that differ by case or punctuation are one
genre (jazz+funk, jazz-funk); other and unknown are not genres.
- Made for the listener: Your favourites, Rediscover (liked artists not played for a
month; both need history) and New to you (songs never played).
- A mix has up to 50 songs, at most 3 by one artist, never two of an artist in a row,
picked by weighted chance (artist preference, never-played, less if played lately;
skipped lately and avoided artists are out). The seed is user, mix, day and variant,
so a mix is steady for a day and another is one variant away.
- Worked out when asked, from the listener's own history and the genre tags; nothing is
stored and nothing leaves the server.
internal/mixes holds the selection and the service, internal/storage/mixes.go the
queries (SQLite and PostgreSQL). Spec, contract test, README and web types follow;
the web app comes separately.
Closes#95
A Playlists page lists the user's playlists and makes a new one. A playlist has its
own page: play it (songs that are not available are skipped), rename it, delete it
after a confirmation, move a song up or down and take one out. Every song of a list
gets an Add to playlist button that opens a window to choose a playlist or make a
new one, and says what it did.
TrackRow keeps the song as one button and takes extra buttons at its end, so the
add button is not inside the button that plays. A long playlist is read whole,
page after page. English and Italian texts, error texts for the new codes.
Closes#92
playlist_items.track_id is ON DELETE CASCADE (added by the playlists work on main), so deleting a
duplicate would have taken the song out of every user's playlists, as it would have done with the
listens. RemoveCopy now repoints the items first, in the same transaction.
Also merges main (playlists) into the branch.
Each user can make playlists: a name (1 to 200 characters), an optional
description and an ordered list of tracks, where a track may appear more than once
(each place has its own item id). The API makes, lists, renames and deletes
playlists, and adds tracks (at the end or at a position), lists them, moves one and
removes one; positions count from 0 without gaps.
A playlist is its owner's alone: somebody else's answers playlist_not_found (404),
as one that does not exist. A track whose files were deleted or moved stays in the
playlist and is listed with available false. Deleting a user deletes their
playlists. Limits: 1,000 playlists per user, 10,000 tracks per playlist, 500 tracks
per request.
Changes of one playlist are serialised by locking its row, so two requests at once
keep the positions unique on PostgreSQL too. Migration 0019, OpenAPI spec, contract
test, README and the web API types follow. The web app comes separately.
Closes#89
Found reviewing the branch: a request cancelled between deleting the file and updating the index left a track whose file does not exist. The index step no longer depends on the request's context. Test first: it cancels the request at exactly that moment.
README: the Duplicates feature, the new setting and its environment variable, the endpoints, and the two places that promised the music folders are never written. The example config and the compose file say how to turn it on and that the folder must then be writable.
Admin only. GET /library/duplicates lists the groups with what is needed to choose and never a path; POST|GET|DELETE /library/duplicates/check is the job that fingerprints the files; POST /library/duplicates/clean deletes the chosen copies of one song, each refusal with its own status and code (deletion_disabled, scan_running, invalid_input, track_not_found, duplicate_changed, library_read_only). In the spec with the contract test, and wired in main, where an info line says when library.allow_delete is on.
Groups come from the index. The check job reads the files of the groups and keeps their SHA-256, skipping a file that changed since the scan. Clean refuses unless deleting is on and no scan runs, works out again that the tracks are copies of one song, checks every file is still what the scan saw (so the last copy is never removed, and a link or a path that leaves the library is refused) before removing any, removes each file through an os.Root, then takes its track out of the index and moves its listens. Tested with real files, including the refusals; three of the safety rules were broken on purpose to see the tests fail.
Candidates are the playable files whose title another file shares. A fingerprint is kept only for the file that was read (same size and date), and a scan that finds the file changed forgets it. Removing a copy moves its listens to the track that stays, because they would be deleted with the track, then removes the track with its file, credits and genres. Tested on SQLite and PostgreSQL.
Many titles end with a note taken from the tags: (disc 2), (bonus disc),
(Deluxe Edition), sometimes cut off by a tag that is too short. A release group is
named after the album, not after a disc, so those albums found nothing.
The lookup now asks for the title as it is first and, only when MusicBrainz has no
sure result, once more without the notes at its end (baseTitle). A group in brackets
with none of the words of a note stays part of the title, and a title that would be
left empty is kept. The rule for accepting a result does not change.
On 25 albums of a real library with such notes, 19 found a cover (none before).
Closes#87
The same song is the same set of main artists, the same title and durations within 3 seconds, every pair within it; a file with no duration or no artist never matches. A group says whether its files were all read and are byte for byte the same. The copy to keep is the lossless one, then the higher bitrate, then the bigger file, then the oldest, whatever the order of the input.
The one setting that lets Sonda delete files from the music folders. Off until the operator turns it on, in the file or with SONDA_ALLOW_DELETE; a value that is not true or false is an error, not ignored.
The covers asked the Cover Art Archive only for albums whose files carried a
release group id. In a library tagged without MusicBrainz, 3,321 of 3,433 albums
have none, so almost no cover could come from the archive.
An album with no id is now looked up in MusicBrainz by the name of its artist and
its title (musicbrainz.Client.FindReleaseGroup), when the musicbrainz provider is
on too. A result is taken only if MusicBrainz scores it 95 or more and its title
and one name of its artist credit equal the ones asked for after folding case,
accents and punctuation; the first such result wins and anything doubtful is no
cover. The id is used for that request and is not stored on the album.
Artists are still never looked up by name. The MusicBrainz and Cover Art Archive
provider texts, the README and the API spec say what is sent.
On 47 albums sampled from a real library, 23 got a cover (all checked as the right
ones by eye or by exact title and artist); a title like 'Live (disc 2)' finds nothing.
Closes#83
make build builds the app and then sondad with -tags webui. The Dockerfile builds the app in a Node stage and embeds it in the Go build, so the image serves it with no Node on the host. README: how to build, what the image holds, the policy of the page, and how to work on the interface with the dev server.
internal/webui carries web/'s build (written to internal/webui/dist by Vite, git-ignored) when sondad is built with -tags webui. Without the tag the app is absent and a plain go build and go test still work with no Node. main wires it into the router.
A GET or HEAD outside /api is a file of the app, or its page for a link inside it (/admin/users). A file that is missing (under assets/ or with an extension) answers 404, never the page, so a missing script is not run as HTML; hidden files and folders are never served; everything else, and any unknown /api address, keeps the JSON 404. The page gets a Content-Security-Policy that allows only its own server (no address, no wildcard, no inline script), nosniff, DENY and no referrer. Hashed assets are cached for good and the page is asked again. With no app in the binary, / says how to build it. The app is an optional dependency, so existing wiring is unchanged.