Add a preview-and-review workflow that matches Hardcover books by cached IDs, ISBNs, slugs, and title/author confidence before updating BookOrbit read status.
Import reading progress only when requested and safe, preserve existing local progress, and keep status imports resilient when progress writes fail.
Closes#394
Use the current book's effective metadata provider configuration to decide which provider ID fields and search providers are available in metadata editing flows.
Preserve explicit empty provider selections so library overrides can remove all provider fields.
Closes#320
Emit transfer events when watcher reconciliation moves a book between libraries. Keep file-as-book and folder-as-book watcher paths consistent across Docker, direct runs, and delayed copy-delete moves.
Closes#386
Send injected empty JSON bodies as Buffer chunks so Nest Fastify parsers can consume them without crashing.
Acknowledge BookOrbit collection item updates locally instead of proxying synthetic tags upstream.
Closes#396
The SeriesMembershipEditor was losing focus on every keystroke when typing into a previously empty series row because the key was bound to the seriesName, causing the input component to remount. This introduces stable row keys to ensure the input retains focus while updating the model.
Closes#370
Expose zero-book first-class metadata entities in browse so admins can review orphaned rows before deleting them.
Keep inline metadata values out of empty filtering because they only exist through book metadata rows.
Closes#370
The central directory entries are now sorted by filename to ensure the correct first image is used as the cover, rather than relying on the order of entries in the ZIP archive.
Closes#378
The GitHub Actions Trivy scan was failing due to CVE-2026-12151, CVE-2026-6734, and CVE-2026-9697 in undici version 7.24.7. It is a transitive dependency of cheerio in the server package. Running pnpm update undici removes the vulnerable version from the lockfile.
Offload large PDF parsing to a worker thread to prevent the Node.js event loop from being blocked. This ensures the application remains responsive during heavy metadata extraction tasks.
Closes#371
Update cover mutation paths to bump the book timestamp used for cover cache busting, and carry per-book collapsed-series cover versions so child covers do not reuse a representative book timestamp.
Introduces a new feature to fetch, parse, and display app release notes
and update highlights directly within the UI.
- Backend: Adds the `release-notes` module to fetch GitHub releases
and parse markdown highlight sections.
- Frontend: Adds the `WhatsNew` dialog and view, including custom
markdown rendering and an image lightbox for update media.
- Tools: Adds validation scripts (`whats-new-lint`, `whats-new-preview`)
to ensure release content parses correctly before publication.
pdf-lib cannot re-encrypt files, so saving an encrypted PDF strips its encryption but writes back still-encrypted byte streams, silently corrupting the file. Skipping the file write entirely prevents this corruption while allowing DB metadata to persist separately.
Fixes#368
Books without a row in user_book_status returned NULL and sorted differently than books explicitly marked as 'unread'. Coalescing the missing status to 'unread' ensures both explicitly and implicitly unread books are grouped and ordered consistently.
Fixes#367
The bulk rename job was incorrectly iterating over every book in the library instead of restricting the operation to books that actually require a rename. This caused unnecessary processing and logged excessive noise. We now compute the preview first and filter the list of candidate book IDs down to those with the 'will_rename' status before executing the rename loop.
Closes#361
Expand coverage across previously untested modules to close gaps identified
in the coverage audit.
BE: entity-manager strategies (inline, junction, author), koreader chapter
extractor, series and achievement repositories.
FE: reader settings/progress composables, notifications, live-scan books,
entity manager, and API wrapper modules (author, statistics, entity-manager).
* feat(metadata): add Aladin Korean book metadata provider
Add support for fetching Korean book metadata from Aladin OpenAPI.
Includes search by title/author/ISBN and lookup by ISBN-10/13.
- Add ALADIN to MetadataProviderKey in @bookorbit/types
- Add AladinProviderConfig to ProviderConfigurations
- Create AladinProvider with IdentifiableProvider interface
- Add mapper for Aladin API response to MetadataCandidate
- Register provider in MetadataFetchModule
- Add provider config merge, validation, and connection test
- Include 24 unit tests (provider + mapper)
* feat(client): add Aladin provider UI integration
* refactor(metadata): use 'Aladin' label and favicon-based icon
* feat(metadata): complete Aladin provider end-to-end wiring
Persist the Aladin ItemId across fetch, scoring, locks, and file metadata round-trip, and update the affected tests.
Co-authored-by: Kyungwook Park <288273+kyungw00k@users.noreply.github.com>
---------
Co-authored-by: neon <18559658+neonsolstice@users.noreply.github.com>
Co-authored-by: Kyungwook Park <288273+kyungw00k@users.noreply.github.com>
This introduces a many-to-many relationship between books and series, resolving the previous limitation where a book could only belong to a single series. It updates the database schema, query builders, server DTOs, and the Vue client interface to manage multiple series memberships per book.
Closes#353
String/math based extraction of floating point series index fraction parts caused floating point precision artifacts. This extracts formatting into a shared utility and uses string splitting to avoid precision issues.
Closes#360
The "Create link" button was missing from the magic links admin tab because the header was hidden. This adds an embedded create action so the button is available when the component is rendered without a header.
Closes#357
Inject an empty JSON object for requests that have an unsupported content type but zero content length. Fastify defaults to rejecting these with a 415 error.
Closes#348
The cover shadow moves to the inner artwork frame. The surface's own overflow:hidden would clip that shadow, so let it render by setting it to visible. The inner frame keeps its own overflow:hidden, so the image stays clipped to its rounded rect.
Closes#298
Adds a display preference to conditionally render book spines on comic
formats. The new setting is defaulted to false. Various book cover
surfaces across the application now respect this preference by checking
if the format is a comic before rendering the spine.
Closes#295
Adds the isUpNext operator to seriesStatus allowing books to be filtered by whether they are the next unread book in a series where the previous book is completed.
Closes#321
Calibre 9.x (opf3) writes provider identifiers as bare `prefix:value` text
inside `<dc:identifier>` instead of using `opf:scheme` attributes. It also
stores custom-column values like page count and subtitle in a
`calibre:user_metadata` JSON blob. This adds parsing for these fallback
formats to correctly extract metadata from Calibre-processed EPUBs.
Closes#341
Introduce devices and rating evaluators backed by a new reading_sessions.source
column, so badges can reward multi-device reading and book ratings. Wire the
source through Kobo and reading-session services and seed the new badges.
Introduces PathService.createDirectory with comprehensive path validation
to prevent directory traversal and symlink attacks. It validates parent paths
against a blocked list, ensures the parent is a real directory, and blocks
unsafe characters in folder names.
Filesystem errors like EEXIST and EACCES are mapped to standard HTTP
exceptions to provide clear feedback to the client.
Closes#332
Rank physical and ebook editions higher than audiobooks in hardcover
metadata mapping. Do not allow audiobooks to inherit page counts from
the base book to prevent incorrect metadata.
Closes#334
The upload paths saved only the aggregate book_metadata duration, leaving book_files.durationSeconds
null so the player summed to zero. Extract and aggregate per-file duration on new uploads and added
audio files, matching the scanner.
Ref #330
The previous edition matching logic often resulted in suboptimal edition assignments, tracking text books against audiobooks or selecting editions with drastically different page counts. A more rigorous matching strategy is required to ensure users get accurate edition metadata from Hardcover.
Closes#333
Fetch requests now await existing promises for in-flight blocks instead of re-triggering and starving the load when crossing block boundaries. The navigation window entries are also preserved during identical structural updates to avoid infinite watcher loops.
Closes#325
The Library, Series, and Authors filter panels were rendered inside the
scrolling <main> element, so opening a panel after scrolling down placed
it above the visible viewport and it appeared to do nothing until the
page was scrolled back to the top. Render each panel as a sibling of
<main> so it stays anchored regardless of scroll position. A height cap
is intentionally avoided because the filter typeaheads use absolutely
positioned dropdowns that an overflow container would clip.
Add structural regression tests for the Series and Authors views that
assert the panel renders outside the scroll container.
Fixes#326
When a metadata provider returns a page count of 0 (e.g. unknown), the frontend would submit 0, causing a validation error (HTTP 400). This change normalizes 0 to null on both the client (before sending) and server (during validation).
Closes#329
Removes the `assertManualUpdateAllowedForLockTransition` guard that incorrectly rejected the legitimate UI flow of unlocking a field, editing its value, and re-locking it before saving.
Closes#328
* feat(metadata): add LubimyCzytac metadata provider
- Add LubimyCzytac scraper and provider implementation
- Add lubimyczytacId field across types, schema, extractors, and metadata service
- Register lubimyczytac provider in module, pipeline, repository, and config service
- Add provider icon, color, link, and UI support in book detail views
- Add migration for lubimyczytac_id column
Closes#304
* fix(books): keep cover card read button opening reader when thumbnail click prefers details
The overlay primary button should always open the reader regardless of
the thumbnailClickAction setting. That setting only controls card click
behaviour.
* fix(books): add explicit read button to hover overlay when thumbnail click prefers details
When thumbnailClickAction is 'details', a dedicated Read button now
appears in the overlay top row so users can still open the reader
directly without changing their preference.
The book dock metadata service now routes audio formats through the audio extractor to extract chapters, duration, narrators, and other embedded tags. This ensures that audiobook metadata is correctly mapped and persisted during the import process.
Closes#322
Mobile keyboards, such as on Android, may not show a functioning
Enter key by default in this input. Adding enterkeyhint="enter"
allows users to properly submit chips like authors in the
metadata editor.
Closes#310
The `orientation: 'any'` setting in the manifest forces the PWA to rotate even if the device is locked. Removing it restores native behavior.
Closes#309
neon
dependabot[bot]
Kyungwook Park
James McNee
Chris Mansell