Two blocking bugs found while reviewing this branch:
1. Compile error: validateURL() was called in fetchAndParse/doDownload (added
while applying Strix's SSRF suggestions) but the function itself was never
committed - only "Add validateURL and isReservedIP helper functions" was
left as a plain-text suggestion with no one-click apply, and it got missed.
Implemented validateURL/isReservedIP plus a safeHTTPTransport whose
DialContext re-resolves and re-checks the target IP at actual connection
time (not just once via a URL pre-check), so a DNS answer that changes
between the check and the request (DNS rebinding) can't reach a reserved
address - this also covers HTTP redirect targets for free, since redirects
reuse the same Transport. Added AllowLoopbackHTTPForTests() so the
existing httptest-based suite (which binds to 127.0.0.1) still passes
without weakening the guard for any other address.
2. Migration boot failure: the podcast migrations were dated 2026-04-27/28
(when the feature was actually developed), but goose.UpContext (as this
project calls it, no WithAllowMissing) hard-errors on any pending
migration older than the DB's already-applied max version. Any install
already past April on current master would fail to start entirely on
upgrade. Renumbered all 5 podcast migrations to 2026-09-02 (after
everything currently on master). This also meant the podcast_* columns
added to the already-shipped uniform_canonical_ids migration's idColumns
were dead code for any install that had already run that migration -
editing an applied migration's Go source doesn't make it re-run. Reverted
that edit and split the podcast id canonicalization into its own,
later migration (20260902000005) that reuses the same buildIDMap/
applyIDMap machinery. Verified both fresh-install and existing-install
upgrade paths end-to-end against real sqlite DBs: no boot error, and
legacy-shaped podcast ids (plus their FK references) get correctly
rewritten to canonical form.
Verified: full build clean, core/podcasts + server/nativeapi + db/migrations
test suites all pass, gofmt clean.
Rebasing onto master's Podcasting-branch-unrelated changes surfaced several
integration gaps the merge conflicts didn't catch:
- conf.Server.DataFolder is now a Dir type, not a string; podcasts.go and
its tests needed .String()/conf.NewDir() at each call site.
- subsonic.New() gained the podcasts.Podcasts parameter; several e2e/unit
test call sites elsewhere in the tree were still passing the old arg count.
- The uniform-canonical-ids migration (dated after our podcast migrations,
so it runs against a schema that already has the podcast tables) didn't
know about the new podcast_* id columns, leaving them unrewritten while
everything else (including podcast_episode.stream_id's matching
media_file.id) got canonicalized.
- opensubsonic_test.go's expected extension count was miscounted during
conflict resolution (9 Podcasting 2.0 extensions, not 7).
- .gitignore's unanchored `podcasts/` entry from an earlier commit
accidentally matched core/podcasts/ (source) in addition to the
downloaded-episode directory; anchored both to their actual paths.
- Parse podcast:images, podcast:funding, podcast:transcript, podcast:chapters,
podcast:soundbite, and podcast:person tags from RSS feeds
- Add DB migration and repositories for podcast images and funding sources
- Expose Podcasting 2.0 fields in Subsonic API responses
- Tag downloaded episodes with genre=Podcast for playlist filtering
- Update mock data store and fix subsonic test compatibility
Adds full support for the Podcasting 2.0 namespace
(https://podcastindex.org/namespace/1.0) across RSS parsing, persistence,
Subsonic API responses, and OpenSubsonic extensions.
- podcast:guid: channel-level UUIDv5 for stable identity across feed URL changes
- podcast:chapters: per-episode chapters URL + MIME type (chaptersUrl in API)
- podcast:transcript: multiple transcripts per episode stored in a dedicated
podcast_transcript table; attributes: url, type, language, rel
- podcast:season: season number + optional name per episode
- podcast:episode: episode number (decimal string) + optional display label
- podcast:person: host/guest entries at both channel and episode level stored in
a dedicated podcast_person table; attributes: name, role, group, img, href;
role defaults to "host" and group defaults to "cast" per spec
- podcast:locked: feed lock flag + optional owner email on channel
- podcast:funding: first funding entry URL + display text on channel
- podcast:medium: content type classification on channel
- podcast:soundbite: startTime (float), duration (float), title per episode
- podcast:updateFrequency: display text + rrule + complete flag on channel
- podcast:podroll: creator-recommended feed list stored in podcast_podroll table
(feedGuid, feedUrl, title, sort_order); returned as podroll[] in GetPodcasts
- podcast:liveItem: live stream detection stored in podcast_live_item table
(one row per channel, unique index); stores status, start/end times,
enclosure URL/type, and contentLink for fallback playback; returned as
liveItem object in GetPodcasts; Upsert preserves created_at on updates
- podcast:podping: usesPodping boolean on channel; RefreshChannels skips
channels with usesPodping=true (they receive updates via Podping WebSocket)
- 20260428000000_add_podcast20.go: ALTER TABLE adds 9 columns to
podcast_channel, 9 columns to podcast_episode; CREATE TABLE
podcast_transcript (episode_id FK, url, mime_type, language, rel) and
podcast_person (no FK constraints — put() serialises "" not NULL)
- 20260428120000_add_podcast_tier3.go: ALTER TABLE adds uses_podping to
podcast_channel; CREATE TABLE podcast_podroll and podcast_live_item
(UNIQUE INDEX on channel_id)
GetPodcasts response (PodcastChannel) gains:
podcastGuid, locked, medium, fundingUrl, fundingText, updateFrequency,
complete, usesPodping, person[], podroll[], liveItem{}
GetPodcastEpisode response (PodcastEpisode) gains:
season, seasonName, episode, episodeDisplay, chaptersUrl, soundbiteStart,
soundbiteDur, transcript[], person[]
GetPodcastEpisode now loads transcripts and persons from their repositories
(previously only read the base episode row).
New OpenSubsonic extensions declared:
podcastChapters, podcastTranscripts, podcastSeason, podcastPerson,
podcastFunding, podcastMedium, podcastPodroll, podcastLiveItem, podcastPodping
- core/podcasts/rss_test.go: 41 new specs covering all namespace tags,
default value handling (role→"host", group→"cast"), backward compatibility
- core/podcasts/podcasts_test.go: 26 new service specs covering AddChannel
field persistence, transcript/person saving, podroll/liveItem saving,
RefreshChannels podping skip behaviour
- persistence/podcast_transcript_repository_test.go: 10 specs
- persistence/podcast_person_repository_test.go: 12 specs
- persistence/podcast_podroll_repository_test.go: 10 specs
- persistence/podcast_live_item_repository_test.go: 8 specs
- server/subsonic/podcasts_test.go: 15 new handler specs for Tier 2 and
Tier 3 fields in GetPodcasts and GetPodcastEpisode responses
Pass the server root context (ctx) to NewPodcastService so that background
download goroutines are tied to the server lifecycle and will be cancelled
on shutdown, matching the pattern used by scanner.New.
Signed-off-by: ji-ho lee <search5@gmail.com>
- Add 30s timeout to episode download HTTP client
- Add 15s timeout to RSS feed fetch HTTP client
- Fix N+1 query in GetAll(withEpisodes): fetch all episodes in a single
query using IN clause via GetByChannels
- Add sanitizeMetadata helper to strip null bytes from ffmpeg tag values
- Add TODO comment on background goroutine context for server shutdown
Signed-off-by: ji-ho lee <search5@gmail.com>
- Add ExistsByURL to MockPodcastChannelRepo
- Add channel mock data to DownloadEpisode error handling and timestamp tests
- Fix DeleteEpisode test to match actual behavior (resets to new status)
- Pass podcasts.Podcasts to subsonic.New in e2e test suite
Signed-off-by: ji-ho lee <search5@gmail.com>
Backend
- Add podcast data model (PodcastChannel, PodcastEpisode) with migrations
- Implement Subsonic API endpoints: getPodcasts, getNewestPodcasts,
createPodcastChannel, refreshPodcasts, deletePodcastChannel,
deletePodcastEpisode, downloadPodcastEpisode, getPodcastEpisode
- Add native REST API endpoints: GET/DELETE /api/podcast,
GET /api/podcast/preview (feed info without creating channel)
- Inject events.Broker into podcast service for SSE support
- Emit PodcastEpisodeProgress SSE events during download (every 512 KB)
and on completion/error with status field
- Use HTTP Content-Length as fallback when RSS feed omits enclosure size
- Write ID3 tags (title, album, genre=Podcast) to downloaded files via
ffmpeg so the library scanner reads correct metadata
- Set MediaFile fields (Title, Album, AlbumID=channelId, AlbumArtist,
Genre) on episode registration
- Add duplicate URL check in AddChannel
- Add ExistsByURL to PodcastChannelRepository
Frontend - Podcast list
- Add grid/list view toggle (Redux podcastViewReducer) matching album list
- New PodcastGridView component with responsive column count (2-6 cols)
- Cover image 100px in table view
- Remove Feed URL column; add inline copy-to-clipboard button
Frontend - Podcast show (episode list)
- Real-time download progress (%) in Status column via SSE, no polling
- Spinner only before first SSE event; N% once data arrives
- Size column removed; Downloading badge replaced with progress
- Completed episodes play on row click; separate play button removed
- Play / Shuffle / Play Next / Add to Queue buttons above episode list
(only shown when completed episodes exist)
- On download completion, reload episodes to obtain streamId for
immediate playback without page refresh
- Clicking album name in AudioTitle navigates to podcast channel page
Frontend - Podcast create
- Full-width URL input with Fetch Feed Info button and Enter key support
- Preview card (cover image, title, episode count, description) before
committing channel creation
- Add Channel button appears only after preview; shows already-registered
message if channel URL exists
Frontend - Playlist
- Album link navigates to podcast channel page for podcast tracks
(identified by genre=Podcast)
- Artist column shows '-' for podcast tracks with empty artist field
Closes#5420
Signed-off-by: ji-ho lee <search5@gmail.com>