Feature Request: Generic Watchfolder Support for Manual Ingestion & Post-Processing (eBooks & Audiobooks)

[pappaq]

Description

Add a Watchfolder / Staging Directory feature to Shelfmark that automatically detects manually added or externally downloaded eBook and audiobook files, processes them through the standard metadata/matching pipeline, and moves them to the appropriate eBook or audiobook library folders.

---

Motivation & Use Case

Currently, Shelfmark handles search, download orchestration, and management exceptionally well when files are retrieved directly through its integrated sources. However, users often have existing collections, external downloads, or niche releases that are obtained manually.

Without an automated ingestion pipeline for manual drops, users must either rely on complex secondary tools (like external scripts or dedicated post-processing daemons) or manually structure, tag, and place files into the media directories.

Having a dedicated watchfolder within Shelfmark would provide a unified entry point for all media—whether downloaded directly via Shelfmark or dropped in manually.

---

Proposed Workflow

1. Directory Monitoring: Shelfmark monitors one or more configurable watch/ or inbox/ directories.
2. File Detection & Stability Check: When new files or folders are detected (e.g., .epub, .pdf, .m4b, .mp3 directories), Shelfmark waits until writing is complete (file lock/stability check).
3. Metadata Matching & Parsing: Shelfmark parses the folder/file names to query metadata providers (e.g., Google Books, Open Library, Audnexus/Audible).
4. Media Type Routing:
   • Text formats (.epub, .pdf, .mobi, etc.) are processed and moved to the configured eBook Output Directory.
   • Audio formats (.m4b, .mp3, .flac, etc.) are processed (and optionally merged/tagged if applicable) and moved to the configured Audiobook Output Directory.
5. Organization & Renaming: Output files are organized and renamed based on the user's configured folder structure patterns (e.g., {Author}/{Title}/{Title}.{ext}).

---

Key Features / Options to Consider

• Separate Inboxes vs. Unified Inbox: Either dedicated folders (e.g., /watch/ebooks and /watch/audiobooks) or a smart auto-detection mechanism based on file extensions.
• Review Queue / Fallback: If metadata cannot be matched automatically with high confidence, place the item in an "Unmatched" UI view for manual resolution rather than silently failing or mislabeling.
• Configurable Polling Interval / inotify: Option to set polling intervals or leverage filesystem event triggers.
• Tagging & Embedded Metadata: Embed matched metadata and cover art into the final files before moving.

---

Benefits

• Streamlines library ingestion into a single workflow.
• Bridges the gap between automated downloads and manual file handling.
• Seamlessly integrates with downstream library servers (such as BookOrbit, Audiobookshelf, Calibre-Web, etc.).

[calibrain]

To be honest, this is really not the goal of shelfmark, and there exists tools to do that such as Calibre-Web-Automated, Grimmory, Bookorbit, Bookshelf, etc.

Shelfmark main goal is "I want this book, get it for me"

I will leave this discussion open, if more people want to chime in and give their views !

[downey-lv]

I agree, this sounds like exactly what CWA does