* feat(opds): add option to skip chapter metadata feed
Introduces a new server configuration `server.opdsSkipChapterMetadataFeed` (default: false).
When enabled, the OPDS chapter feed generates direct acquisition (CBZ download) and streaming (OPDS-PSE) links within the chapter list entries, bypassing the intermediate metadata subsection. This streamlines the user experience and improves compatibility with OPDS clients like KOReader that rely on direct links for automated downloading features.
* fix: lint
* fix(opds): enrich chapter data and refine sync logic for skip-metadata mode
Refines the `opdsSkipChapterMetadataFeed` implementation to ensure necessary data is available for direct links and handles synchronization logic appropriate for a list view.
- **Refactor ChapterForDownload:** Extract `refreshChapterPageList` and `updateChapterPersistence` to allow reusing page count verification logic outside the download flow.
- **Enrich Chapter Repository:** When skipping metadata, asynchronously verify page counts and calculate CBZ file sizes for chapters in the list. This ensures direct stream/download links are valid even if the chapter wasn't previously fully indexed.
- **KoSync Logic:** Implement synchronization logic in `OpdsEntryBuilder`. Since the user cannot be prompted in the chapter list view, `PROMPT` conflicts are explicitly ignored (prioritizing local progress), while updates are applied if non-conflicting.
- **OPDS Attributes:** Add `length` (file size) to acquisition links and ensure download links only appear for actually downloaded chapters.
- **Documentation:** Update `server.conf` description to clarify KoSync behavior in this mode.
* feat(download): improve chapter download filenames
* feat(opds): append language to source names
* feat(opds): handle empty chapter titles
* fix import org.jetbrains.exposed.v1.core.inList
* refactor(opds): reorganize API routes and update facet count calculations based on active filters
- **API Routing & Controllers**: Reorganize OPDS v1.2 route paths into logical groups in `OpdsAPI`. Centralize request filter extraction into `OpdsMangaFilter.fromContext`.
- **Facet Counting**: Extract `Query.applyOpdsMangaFilter` to apply active filters to facet and navigation queries. Pass the active filters to `NavigationRepository` and `MangaRepository` count queries (using `excludeField` to calculate sibling counts). This ensures category, source, language, status, and genre counts (`thr:count`) are accurately computed based on active selections.
- **Pagination**: Add pagination support to computed navigation feeds in `NavigationRepository` ( statuses and content languages).
- **Builders**: Standardize parameter ordering in `FeedBuilderInternal` and `OpdsEntryBuilder` constructors. Simplify pagination and facet link URL generation.
* fix(opds): remove redundant filter logic to avoid duplicate HAVING clauses
Resolve IllegalStateException crash caused by applying content filters twice in MangaRepository. Filtering is now handled exclusively by `applyOpdsMangaFilter`, allowing `applyMangaLibrarySort` to focus solely on ordering operations.
* revert(download): restore original CBZ filename scheme
* refactor(opds): simplify persistence updates and clean up chapter mapping
- Simplify page count and download checks in ChapterForDownload
- Clean up enriched chapter mapping in ChapterRepository to improve readability
* fix(opds): retrieve chapter archive size without leaving stream open
* perf(opds): avoid redundant DB query when refreshing chapter page list
The returned result rows of the inserted chapters did not have the up-to-date "last_modified_at".
This caused "downloadNewChapters" to not be able to correctly detect unread chapters. it included the newly inserted ones, leading to exiting early due to having unread chapters.
Regression 811e15162bfixes#2097
* Implement SyncYomi
* Add ability to select what to sync
* Properly fix default category bug
* Add periodic sync
* Add PostgreSQL support
* Deschedule previous task
* Check if SyncYomi is enabled in syncData function
* Don't allow multiple syncs at the same time
* Convert SyncYomiSyncService to object
* Make startSync non-suspend
* Return a result from startSync
* Sync before library update
* Improvements
* Use NetworkHelper client
* Lint
* Use measureTime
* Database improvements
- Move entire sync operation into a single transaction
- Stop loading all manga to memory
* Revert "Database improvements"
This reverts commit bee8d214c3.
* Actual database improvements
* Remove runBlocking
* Remove title check
* Update updateNonFavorites function
* Update timeout code
* Improve PostgreSQL query
* Create lastSyncState variable
* Create lastSyncStatus query
* Convert lastSyncState to StateFlow
* Create lastSyncStatusChange subscription
* Replace backupRestoreStatus with backupRestoreId
* Add startDate and endDate
* Add logs for sync start and end
* Handle all errors in syncData
* Change category restore function to match Mihon's behavior
* Fix comment
* Remove duplicate BackupMangaHandler.backup call
* Remove duplicated log
* Rename subscription to syncStatusChanged
* Use same flags for restoring
* Update syncInterval config to use DurationSetting
* Update sync scheduling logic
* Reorder conditions to reduce database calls
* Prevent deleted ghost chapters from reappearing during sync
jobobby04/TachiyomiSY#1575
* Improve sync merging categories
jobobby04/TachiyomiSY#1559
* Make columns not null
* Improve H2 triggers
* Add documentation
* Fix renaming manga download dir
* Simplify manga download dir rename function
---------
Co-authored-by: Mitchell Syer <Syer10@users.noreply.github.com>
The deletion of chapter data was done in its own transaction. Thus, when the update or insertion failed later on, the deletion was not rolled back
fixes#2031
Co-authored-by: Mitchell Syer <Syer10@users.noreply.github.com>
* Reset update-flag on uninstall
If there is an update available when the extension is uninstalled, the
table will still have the update flag, which makes no sense if it is not
installed.
Example:
```
{
"pkgName": "eu.kanade.tachiyomi.extension.en.comix",
"name": "Comix",
"lang": "en",
"versionCode": 20,
"versionName": "1.4.20",
"iconUrl": "/api/v1/extension/icon/tachiyomi-en.comix-v1.4.20.apk",
"repo": "<hidden>",
"isNsfw": true,
"isInstalled": false,
"isObsolete": false,
"hasUpdate": true,
"__typename": "ExtensionType"
},
```
* Update changelog
* fix: Do not return `inputStream` from conversion
The returned value must be owned, since the caller closes the input
stream on success
* fix: Assume a conversion error consumes the input stream
e.g. converting an ARGB png to jpeg will throw "bogus colorspace", but
only after the inputstream is consumed. so in case of an exception, we
have to assume that the stream is broken and re-open the page from cache
* Remove koreader-sync credentials from config
These are supposed to be set via the login/logout mutations and are not meant to be set manually by the user. Thus, they are not really settings and do not belong to the config
* Reduce log levels of KoreaderSyncService
* Add backup flags to auto backups
* Mark ServerConfig properties as deprecated
---------
Co-authored-by: Mitchell Syer <Syer10@users.noreply.github.com>
* Extract global metadata backup logic into BackupGlobalMetaHandler
* Extract category backup logic into BackupCategoryHandler
* Extract source backup logic into BackupSourceHandler
* Extract manga backup logic into BackupMangaHandler
* Add way to exclude settings from backups
* Exclude flaresolverrEnabled
* Exclude usernames/passwords
* Exclude writing deprecated settings to the backup
* Exclude AuthMode
* Skip thumbnail download for local manga sources
Local manga sources do not require downloading thumbnails as they are stored locally.
* Always update local source manga info when browsing
When making changes to an in library local source manga, a refresh from the source was required to get the latest data.
From a user perspective, this is unexpected behavior that looks like a bug.
If, for example, the thumbnail file extension got changed, the file could not be found anymore and an error was shown in the client. To fix this, a manga refresh was required.
* refactor(kosync): introduce differentiated sync strategies
Replaces the single `koreaderSyncStrategy` setting with `koreaderSyncStrategyForward` and `koreaderSyncStrategyBackward`. This allows users to define distinct conflict resolution behaviors based on whether the remote progress is newer or older than the local progress.
The `KoreaderSyncStrategy` enum has been simplified to `KoreaderSyncConflictStrategy` with four clear options: `PROMPT`, `KEEP_LOCAL`, `KEEP_REMOTE`, and `DISABLED`. The ambiguous `SILENT` option is removed, as its behavior is now implicitly covered by selecting `KEEP_REMOTE` for forward syncs and `KEEP_LOCAL` for backward syncs.
The legacy `koreaderSyncStrategy` setting is now deprecated and is seamlessly migrated to the new dual-strategy system using `MigratedConfigValue`, ensuring backward compatibility for existing user configurations.
* fix(kosync): correct proto numbers and setting order for sync strategies
* fix(kosync): proto number 78 to 68
* fix(server): migrate KOReader sync strategy during settings cleanup
Add migration logic to convert the old `server.koreaderSyncStrategy` key
into the new `server.koreaderSyncStrategyForward` and
`server.koreaderSyncStrategyBackward` keys during server setup.
* Cleanup graphql setting mutation
* Validate values read from config
* Generate server-reference.conf files from ServerConfig
* Remove unnecessary enum value handling in config value update
Commit df0078b725 introduced the usage of config4k, which handles enums automatically. Thus, this handling is outdated and not needed anymore
* Generate gql SettingsType from ServerConfig
* Extract settings backup logic
* Generate settings backup files
* Move "group" arg to second position
To make it easier to detect and have it at the same position consistently for all settings.
* Remove setting generation from compilation
* Extract setting generation code into new module
* Extract pure setting generation code into new module
* Remove generated settings files from src tree
* Force each setting to set a default value
When generating a hash on-the-fly for the binary checksum method, the code was performing a full MD5 hash of the entire file stream. This was incorrect as the KOReader binary method expects a specific partial hash (reading small chunks at different offsets).
This change ensures that when an in-memory CBZ is created, it is first written to a temporary file. Then, the correct `KoreaderHelper.hashContents()` function is used on that file to generate the partial hash, matching KOReader's logic. The temporary file is deleted immediately after.
* fix(archive): unify CBZ generation to produce deterministic archives
Previously, CBZ files generated on-the-fly (`FolderProvider`) had a different hash than those created directly (`ArchiveProvider`), even with identical content. This inconsistency was caused by using two different ZIP libraries (`java.util.zip` vs. `org.apache.commons.compress`) and not normalizing file metadata.
This inconsistent hashing breaks binary-based synchronization with external services like KOReader Sync Server, as the same chapter could be identified as a different file on each generation.
This change ensures CBZ generation is fully deterministic by:
- Unifying both providers to use `org.apache.commons.compress`.
- Setting a fixed epoch timestamp (`time = 0L`) for all ZIP entries.
- Explicitly setting the compression method and level to `DEFLATED` with default compression.
This guarantees that a CBZ file for a given chapter will always have the same hash, regardless of how it's generated, resolving synchronization issues.
* feat(kosync): lazily generate and cache CBZ hashes for sync
Previously, KOReader progress sync in binary mode was limited to chapters explicitly downloaded as CBZ files. Chapters stored as folders lacked a hash, preventing them from being synced.
With the recent move to deterministic CBZ generation, it's now possible to create a consistent hash for any downloaded chapter on-the-fly.
This commit enhances the `getOrGenerateChapterHash` function to act as a central point for hash management. If a hash is requested for a downloaded chapter that doesn't have one cached in the database:
1. It generates the CBZ archive in-memory from the downloaded folder or existing CBZ using `ChapterDownloadHelper.getAsArchiveStream()`.
2. It calculates the deterministic hash of the generated archive content.
3. It saves this hash to the `koreader_hash` column in the `Chapter` table for future use.
The cached hash is cleared when the chapter download is deleted, ensuring hashes are only tracked for available content.
This change transparently extends Koreader Sync compatibility to all downloaded chapters, regardless of their storage format, without requiring users to pre-convert their library to CBZ.
* fix: rename getAsArchiveStream to getArchiveStreamWithSize
* feat(opds): Enhance KOSync conflict handling and reliability
This commit introduces several improvements to the KOSync integration within the OPDS feed, focusing on fixing bugs, improving network stability, and enhancing user feedback during synchronization conflicts.
- fix(sync): Corrects a KOSync JSON deserialization issue by mapping the `updated_at` field from the server response to the `timestamp` property in the client's data model. This resolves a critical bug where remote progress was being ignored.
- fix(sync): Adds a `Connection: close` header to all KOSync API requests. This prevents `unexpected end of stream` errors by ensuring a fresh connection is used, improving network reliability.
- feat(opds): Resolve sync conflicts by generating separate OPDS entries for local and remote progress. This aligns with the OPDS-PSE specification's implicit design of one stream link per entry. Instead of incorrectly adding multiple links to a single entry, the feed now presents two distinct, clearly labeled entries, allowing users to choose their desired reading position from compatible clients.
- chore(sync): Adds detailed debug logging for KOSync `GET` and `PUT` requests, including request URLs, sent data, and received responses. This improves traceability and makes debugging future issues significantly easier.
* change synced icon
* unnecessary comments removed
Previously, handling a HEAD request on the chapter download endpoint was inefficient as it triggered the full CBZ file generation process in-memory just to retrieve metadata like Content-Length and Content-Disposition. This caused unnecessary latency especially for OPDS clients.
This commit introduces a separate, lightweight path for HEAD requests.
- A new `getCbzMetadataForDownload` method is added to `ChapterDownloadHelper` to calculate the filename and file size without generating an archive stream.
- The `ChaptersFilesProvider` interface is updated with a `getArchiveSize()` method, implemented by both `ArchiveProvider` and `FolderProvider`, to retrieve the total size.
- The `MangaController` now differentiates between GET and HEAD methods, invoking the appropriate helper to ensure HEAD requests are served instantly with only the required metadata.
