feat(dvr)!: unify live and VOD/series recording into one recoverable DVR
Merge fix/merge_live_and_vod_recording into develop.
- One recording queue for scheduled live captures and VOD/series transfers,
with a shared transition graph, provider-slot scheduling, resumable HTTP
transfers (strong ETag / Last-Modified validation) and ffmpeg live capture.
- Crash-recoverable persistence: B+Tree repository with a journal-first
recovery history; one file can be shared by several library entries.
- Admission checks quota and disk on request, at transfer start (real size)
and before an entry attaches to a finished file; idempotent requests.
- REST is command-only; state reaches clients via owner-filtered WebSocket
snapshots.
- Deleting a recording removes the file once no entry holds it; removing an
entry keeps a finished file and cleans up orphaned partials.
- User ids derive from the configured username (web:/api:); the identity
registry is gone.
- Spanish locale added; ru/ar completed; locale parity test.
BREAKING CHANGE: `video.download` is now `video.recording` (nested
`recording` keys move up, `download_priority`/`recording_priority` become
`priority`); old configs fail to load with a migration hint.
BREAKING CHANGE: `/api/v1/file/download*` and `/file/record` are removed in
favour of `/api/v1/recording/*`.
BREAKING CHANGE: `download.*` and `recording.write` permissions are replaced
by `recording.read|create|manage|delete`; tokens from before the change must
re-authenticate.
BREAKING CHANGE: the old recording queue does not migrate; existing
`identity_registry.json` is ignored.
* **New Features**
* Added a setting to limit Flussonic HLS catch-up archive duration. Valid values range from 1 second to 7 days; the default is 4 hours.
* Bounded archive requests are capped at the configured limit, while shorter valid requests retain their requested duration. Changes apply after a playlist refresh.
* Added a dashboard credits card with app and TMDB attribution.
* **Bug Fixes**
* Open filter dialogs now render in the document body when available, helping them display correctly.
* **Documentation**
* Documented the archive-duration setting and bounded-archive behavior.
New Features
Added target-specific bouquet filtering with whitelist and blacklist modes for Live, VOD, and Series groups.
Added an in-context bouquet editor with search, selection controls, status indicators, reset, and streaming previews.
Targets are managed by name, with saved selections available through the Source Editor.
Improved stream alias resolution and provider failover behavior.
Bug Fixes
Existing playlists are preserved when refreshes produce no usable items.
Empty playlists are no longer published.
Configuration changes now recover more safely from persistence failures.
Documentation
Updated feature and configuration documentation and added translations.
* Fix M3U alias credential rewriting on provider fallback
Rewrite opaque authentication query parameters when stream allocation
switches from the primary M3U provider to an alias.
Support unambiguous cross-key mappings such as token to api_key while
preserving unrelated query parameters and rejecting ambiguous mappings.
Add regression coverage for URL rewriting and provider allocation.
New Features
Log consoles now support consistent theming across all available visual themes, including severity and target colors.
Filtering and actions use shared controls with improved accessibility hints and visual states.
Logs are expanded by default, with playlist updates displayed under a dedicated “Playlist Update” section.
Bug Fixes
Improved radio-button group borders and selected-state layering.
Refined stats panel sizing and styling for more consistent layouts.
New Features
- Added first-class Stalker/Ministra portal support, including live, VOD, series, playback-link refresh, and bulk EPG ingestion.
- Added Stalker configuration and batch/CSV support in the Web UI.
- Added playlist preview and playback handling for Stalker inputs, with support for partial refreshes.
* catchup fixes
* feat(frontend): bookmarkable views via URL hash deep linking
Persist the active view to the URL hash so views are bookmarkable and
survive a page refresh. Restore the view from the hash on load, sync the
hash on navigation, and handle browser back/forward via a hashchange listener.
* feat(dev): add dev container for reproducible development
* feat(frontend): interactive multi-series metric sparklines on stats cards
* feat(frontend): aggregate status health banner with capacity breakdown
* feat(frontend): add guided empty states with hints
* chore(devcontainer): add gh CLI, dev tooling, cache volumes, and tasks
- Install GitHub CLI, git, build-essential, clang, lld, mold, jq
- Add cargo-watch, cargo-llvm-cov, cargo-deny, cargo-machete
- Persist cargo registry, target, and gh config via named volumes
- Fix volume ownership for the dev user in post-create
* chore(vscode): add tasks and launch configs for dev workflow
- tasks.json: backend/frontend dev servers, test, lint, fmt, coverage, deps checks, docs
- launch.json: CodeLLDB configs for backend and workspace tests
* a11y: add accessible labels to inputs and collapsed sidebar buttons
- Input: add aria_label prop, falling back to placeholder when no visible label
- Sidebar: pass translated hint + aria_label to collapsed-mode icon buttons
* feat(frontend): warn on unsaved config changes before page unload
* feat(frontend): handle session expiry with client-side logout
Schedule a client-side logout when the JWT expires, showing a
notification and returning the user to the login screen instead of
silently failing with 401s.
* chore(vscode): update tasks.json
* New Features
Session expiration detection with proactive logout warning
Real-time health status banner showing connectivity and provider status
Live metric sparklines for CPU, memory, and network usage
Deep-linkable views via URL hash with back/forward browser support
Enhanced empty states with guided messages across app sections
Improvements
Faster system metrics sampling (2-second intervals)
Archive and catchup-aware EPG handling for accurate program guides
* New Features
Session expiration detection with proactive logout warning
Real-time health status banner showing connectivity and provider status
Live metric sparklines for CPU, memory, and network usage
Deep-linkable views via URL hash with back/forward browser support
Enhanced empty states with guided messages across app sections
Improvements
Faster system metrics sampling (2-second intervals)
Archive and catchup-aware EPG handling for accurate program guides
* Sorce ordinal by category and channel index
* Parsing UserApiRequest from request body or query
* CSS stylings
* media tools refactoring
* xtream info parsing now number/string fields
* Fixed issues with probe and resolve tasks
* hls and catchup session fixes
* fixed problem with grace and shared stream
- added fix for missing episode title
* keep user table info in frontend when switched to custom video stream
* Fixed problems caused by yew upgrade because of UseStateHandle and UseReducerHandle PartialEq change.
* reformatted
* Optimized build
* timeout secs for custom video response
* init custom video timeout on mode switch
* terminate immediately when custom video buffer is missing
* add provisioning missing-custom timeout regression test
* add missing-custom termination tests for all custom modes
* fix ts continuity handling for discontinuity and adaptation-only packets
* fix ts loop duration estimation to use min/max timestamps
* harden pes timestamp patching and normalize pid cc sequence
* Resource ffmpeg creation without B-Frames and short GOPs
* Discontinuity patch for transport stream buffer
* grace hold stream is now default true
* streaming fixes
B+ tree now supports a multi-block layout to handle records that exceed a single block.
Additionally, a larger safety factor was added to the overhead calculation to reduce the risk of data overflow.
Added in-place update for values if they fit in the block without rewriting the whole tree structure.
Implemented an LRU Block Cache (Node Cache) for the query engine.
Implemented zero copy key lookup
Added --dbv for printing id_mapping.db files
Increased initial load time with mmap
* Season Order fix
* fixed episode id problem
* fixed path prefix for user bouquet editor
* fixed css styling for mobile size
* fixed disabled input config
* cargo clippy fixes
An optimization has been introduced to reduce memory consumption during playlist updates.
Overview
Previously, provider playlists were fully loaded into memory during the update process. For large playlists (e.g. hundreds of thousands of entries or multiple providers), this could result in significant RAM usage.
With the new implementation, it is now possible to optionally read provider playlists directly from disk instead of keeping them entirely in memory.
How It Works
- users can configure whether:
- Provider playlists are loaded into memory (previous behavior), or
- Provider playlists are streamed to/from disk to minimize RAM usage.
Processing remains sequential and batch-based, ensuring identical functional behavior.
This approach significantly lowers peak memory usage, especially on systems with limited resources.
Benefits
- Reduced peak RAM consumption during playlist updates
- Better scalability for large playlists and multiple providers
- Full backward compatibility
Trade-offs / Drawbacks
- Increased processing time due to reduced in-memory caching
- Higher disk I/O usage, especially for large or fragmented playlists
- Performance depends more strongly on disk speed (Nvme SSD, HDD)
- Slightly increased CPU overhead due to repeated parsing and deserialization
- Not optimal for environments where fast updates are more important than memory usage
Recommendation
- Use in-memory mode for systems with sufficient RAM and a focus on update speed
- Use disk-based mode for large playlists, multiple providers, or memory-constrained environments
