v1.0.2

Bug-fix release.

Fix

PWA / web clients: library endpoints appeared empty — the CORS middleware only emitted Access-Control-Allow-Origin when the API key arrived via the X-Api-Key header. Clients that pass the key as ?apikey= in the URL (which is what mobile PWAs and any browser-side fetch() does to avoid triggering a preflight) had their responses silently dropped by the browser. CORS now recognizes both forms.

Regression tests added covering header, query-param, and unauthenticated cross-origin.

CI hardening (from #12)

• go vet, staticcheck, govulncheck, docker build now run on every PR
• Bumped to Go 1.25.x (latest patch) for stdlib CVE fixes
• +4 unit tests covering edge cases in the v1.0.1 fixes (short files, weird ZIP signatures, rename collisions, mirror-progress ordering)

Docker image

docker pull ghcr.io/jeremiahm37/librarr:v1.0.2

v1.0.1

Bug-fix release addressing the two issues reported by @cappdom.

Fixes

#7 — Downloads/Searches not working

When the primary libgen mirror (libgen.li) is down or returns 5xx, downloads now automatically fail over to alternate mirrors: libgen.la → libgen.bz → libgen.gl → libgen.vg. Previously a single outage on libgen.li broke all Anna's Archive downloads.

Also improved Prowlarr error messages: when Prowlarr returns HTML (wrong API key, reverse proxy intercept, Authelia login page) the error now clearly identifies the cause instead of a cryptic invalid character '<' JSON-decode error.

#8 — Anna's Archive format mismatch

Files advertised as EPUB that are actually PDF (or vice-versa) are now correctly detected from their magic bytes (%PDF-, PK\x03\x04, Rar!, BOOK, etc.) and saved with the correct extension. Previously a PDF served under an EPUB link would save as .epub and fail EPUB verification.

Testing

• 26 new unit tests covering mirror fallback (5xx, network error, missing link, all-down) and magic-byte detection (PDF / EPUB / MOBI / CBR / unknown)
• Verified end-to-end against the exact MD5 from issue #8 (461cd61c...) — file correctly detected as PDF and renamed
• All pre-existing tests still pass

Docker image

docker pull ghcr.io/jeremiahm37/librarr:v1.0.1
# or
docker pull ghcr.io/jeremiahm37/librarr:latest

v1.0.0

Librarr v1.0.0 -- The Missing *arr for Books

First stable release. Full-featured self-hosted book, audiobook, and manga search and download manager.

Highlights

• 13 search sources with confidence scoring (0-100 with breakdown)
• Quality profiles (format ranking, auto-upgrade) and release profiles (preferred/excluded words)
• Blocklist with auto-add on download failure
• Request workflow with notifications (pending, approved, searching, downloading, completed)
• 4 download clients: qBittorrent, Transmission, Deluge, SABnzbd
• Author monitoring and series auto-complete (detect gaps, search missing volumes)
• Reading history with stats
• Tags for library organization
• Discord and generic webhook notifications
• Goodreads and StoryGraph CSV import
• Library import/export (JSON/CSV) and database backup/restore
• Scheduled wishlist searches with auto-download
• Torznab/Newznab API (add as indexer in Prowlarr/Readarr)
• OPDS 1.2 feed for e-readers
• Multi-user auth with TOTP 2FA and OIDC/SSO support
• Rename on import (configurable pattern)
• Auto-import pipeline to Calibre, Audiobookshelf, Kavita, Komga
• Admin dashboard with source health, activity log, bulk operations
• Prometheus metrics
• Mobile-responsive dark UI
• Single ~17MB Go binary, no runtime dependencies

Full Changelog

https://github.com/JeremiahM37/librarr/commits/v1.0.0

v0.9.0

Changelog

• 7ab1761 Add AI monitor feature with auto-healing and UI tab
• ef19ed5 Add Gutenberg, Open Library, and Librivox search sources
• a3d9d77 Add Proper Authentication
• f7428fc Add Standard Ebooks source, stability fixes, improved Docker setup
• cebaa15 Add badges, CONTRIBUTING.md, improve README discoverability
• b75dcde Add book pipeline features and reliability improvements
• 9efe328 Add comprehensive README with features, API docs, and legal disclaimer
• af2b429 Add comprehensive unit tests — 448 tests across 23 packages
• ebdae7a Add development guidelines and fix search filter (60% word overlap)
• c3c249e Add manga/webcomic search and download support
• 48cfe51 Add multi-user auth, TOTP 2FA, and OIDC/SSO support
• e5f7f92 Add plugin source system for extensible search sources
• 4a9a82a Add post-processing pipeline with multi-library import and tracking
• 879aec1 Add request workflow, metadata enrichment, scoring, admin dashboard, uploads, Deluge + Transmission
• 6920463 Add series management, download reliability, and auto-classification
• 121860b Add settings UI, connection testing, and persistent downloads
• 69520a4 Clean up pipeline: remove dead code, fix token expiry, parallelize dupe checks
• bd465c3 Fix manga pipeline: use correct job._job_id attr and app.library for DB; update README with manga features
• 88b752b Harden runtime readiness and add backup exports
• 79b42ca Improve README — highlight killer features, add request workflow and scoring
• 8f28473 Initial release: self-hosted book search and download manager
• 6bfebc0 Merge PR #1: qBit ban detection, audiobook pipeline, auth, magnet fix
• f33edee Refactor app into modular services and blueprints
• 9d2b148 Rewrite in Go — 17MB static binary, 13 search sources, Torznab API
• 49c2cf3 qBittorrent integration, magnet link resolution & Add audiobook torrent post-processing