How well does this work with Readarr?

[devanteweary]

I just installed this last night and I love the work put into it!

However, unless I'm mistaken, this doesn't really work with Readarr's function as a book database, does it?
Since how you add books is the ingest folder and that removes whatever you put into it.

Or is there a way to use both in conjunction with each other?

Thank you!

[cbrherms]

I was just wondering the same thing while looking in to setting up some sort of book automation. Would be interesting to hear what peoples workflows are like.

[LowerHater]

Going to add my voice here as someone trying to integrate this with Readarr.

[vhsdream]

I can think of a couple of ways to do this:

1. Use the 'Custom Script' option in Readarr's 'Connect' settings to run a script that copies the file to the ingest folder when it is added to Readarr.
2. Or, from the CWA side, use a bash script to watch the Readarr library and have it copy the file to the ingest folder when it detects a new file. The script below is a modified version of the Calibre-Web import script but instead of running calibredb it just copies the files to the CWA ingest folder:

#!/usr/bin/env bash

# Folder to monitor
WATCH_FOLDER="/media/Readarr_library"
echo "Watching folder: $WATCH_FOLDER"

# CWA Ingest Folder Path
CWA_INGEST="/opt/cwa-book-ingest"
echo "CWA Ingest Folder: $CWA_INGEST"

# Monitor the folder for new files
inotifywait -m -r --format="%e %w%f" -e close_write -e moved_to "$WATCH_FOLDER" |
while read -r events filepath; do
        echo "New file detected: $filepath"
        cp -r "$filepath" "$CWA_INGEST"
done

[holocronology]

Do new books need to go to the ingest folder?

What if they just went to the "root folder" in Readarr which is where your library is located?

[holocronology]

Alternatively, you could configure your downloader (at least for usenet via nzb/sabnzbd) to dump downloads into the ingest folder and then CWA would ingest and move to your root folder (aka library folder).

[atabbott]

@vhsdream I really liked your first suggestion to use a custom script triggered by readarr. The suggestion inspired me to write the below script.

I have it set up in readarr to trigger on release import and on upgrade. When that happens it will copy the ebook file to the specified cwa ingest folder. This was only really made possible as I discovered readarr provides a environment variable with the full paths for imported or upgraded books as readarr_addedbookpaths. Amazingly readarr_addedbookpaths is undocumented on the readarr wiki.

Anyway I hope some people find this script useful. It has massively helped me.

Note: I believe readarr_addedbookpaths will also provide a list of paths if a collection is imported. However I haven't been able to test this, but I have tried to handle it by assuming | would be used as the separator. If anyone finds that to be wrong just update the separator used (and let me know if possible).

#!/usr/bin/env bash

# Log file for debugging
LOGFILE="/config/logs/readarr-book-copy.log" # path to your readarr log folder

# Log all environment variables (useful for debugging Readarr's behavior)
# echo "$(date): Script environment variables:" >> "$LOGFILE"
# set >> "$LOGFILE"

# Exit early if READARR_EVENTTYPE is "Test"
if [ "$readarr_eventtype" == "Test" ]; then
    echo "$(date): INFO - Received 'Test' event from Readarr, exiting script early." >> "$LOGFILE"
    exit 0
fi

# Readarr environment variables
BOOK_PATHS="$readarr_addedbookpaths"
DEST_DIR="/media/book-ingest" # Replace with the path to your book ingest folder

# Check if BOOK_PATHS is empty
if [ -z "$BOOK_PATHS" ]; then
    echo "$(date): ERROR - No book paths found in READARR_ADDEDBOOKPATHS." >> "$LOGFILE"
    exit 1
fi

# Process each book path (assuming | is the separator)
IFS='|' read -ra BOOK_ARRAY <<< "$BOOK_PATHS"

for BOOK_PATH in "${BOOK_ARRAY[@]}"; do
    # Trim leading/trailing spaces (just in case)
    BOOK_PATH=$(echo "$BOOK_PATH" | xargs)

    # Ensure the book file exists
    if [ ! -f "$BOOK_PATH" ]; then
        echo "$(date): ERROR - Book not found: $BOOK_PATH" >> "$LOGFILE"
        continue
    fi

    # Extract the filename
    BOOK_FILENAME=$(basename "$BOOK_PATH")

    # Define the destination file path
    DEST_BOOK="$DEST_DIR/$BOOK_FILENAME"

    # Copy the book
    cp -v "$BOOK_PATH" "$DEST_BOOK" >> "$LOGFILE" 2>&1

    # Check if the copy was successful
    if [ $? -eq 0 ]; then
        echo "$(date): SUCCESS - Copied '$BOOK_PATH' to '$DEST_BOOK'" >> "$LOGFILE"
    else
        echo "$(date): ERROR - Failed to copy '$BOOK_PATH'" >> "$LOGFILE"
    fi
done

exit 0

[devanteweary]

Won't any solution where Readarr moves it to the ingest folder still not solve the issue of the management part of Readarr?

Sure it will move it to the CWA ingest folder, but then CWA will ingest it and Readarr will not longer see it and keep track of what you have.

[atabbott]

This creates a copy in the CWA ingest folder, rather than moving it. The copy readarr has available to it remains. This means readarr still keeps track of your library and what you have but it is replicated in CWA.

This does mean you have 2 copies of each book, or if set up correctly one copy with hardlinks (a single copy with 2 refrences to it) so it doesn't solve the issue completly, but ebooks are small enough that I would never delete copies, so I don't really see this as an issue for me.

[devanteweary]

Gotcha. I guess my question would naturally be why is CWA doing it this way with its own database?

Currently my setup is Readarr organizes the books, Calibre does whatever it does (databases those books?), and Calibre-Web / Kavita (I have both setup) presents those existing book files in a nice UI. No files copied/duplicated.

I'm sure there's some reason CWA did it this way. Just wondering if the benefit is something I'm overlooking!

[mathisgauthey]

I see Readarr is capable to send to a Calibre content server. Wouldn't it be possible for it to use the database on Calibre-Web-Automated ?