This is a set of script to help manage and convert comics file along with calibre integration via the content server.
- Read EPUB, AZW3, CBZ, PDF file
- Write EPUB, CBZ, PDF file
- Manage metadata and table of contents across all formats
- Integrated with calibre via content server (NOT a plugin)
The program require Python, probably 3.10+ (I don't know which version is actually required). ImageMagick is required for PDF input (not required for PDF output)
This tools is intended for command line usage.
Download and/or clone the repository and use pip to install dependencies.
$ pip install -r requirements.txt
Use the script-comic.py.
script-comic.py [--width=WIDTH] [--height=HEIGHT] [--rtl] input output
Supported input: folder, zip, cbz, pdf, epub, azw3
Supported output: cbz, zip, pdf, epub
First, make sure to start the calibre content server. Then use the script-calibre-comic.py.
script-calibre-comic.py [--library=LIBRARY_PATH] [--format=FORMAT] [--rtl]
[--no-process] [--width=WIDTH] [--height=HEIGHT]
[--quality=60] [ids [ids [...]]]
--library=URL Specified library path to calibre content server to pass to
calibredb tool.
--rtl Specified that the comic is read from left to right.
Only apply to CBZ input, otherwise the direction
is read from the source file.
--format=FORMAT Output format. PDF or CBZ or EPUB (default)
--no-process Disable image processing
--width, --height Target image width. Default is 1404 by 1872 (7.8" 300ppi).
Only if image processing is enabled.
--quality=QUALITY JPEG quality to save. Default 60. [1-100]
Only if image processing is enabled.
ids Calibre's book id to convert.
This script will prefer format in the order of
AZW3, CBZ, PDF, EPUB as the input.
This is a simple tool to help adding ComicInfo.xml metadata to the CBZ file. Cannot be run directly -- require a local server.
Several image processing are performed on the comic. This can be adjusted manually in the code, but currently it cannot be adjusted from the provided script.
- Heuristic merge of page spread.
- The program calculate contrast at the edge of each spread to decide if it is actually a full spread.
- Margin cropping
- Crop empty border.
- Gamma correction and dithering
- Apply gamma correction and dither the image to 4-bit greyscale.
- Page spread split
Implementation detail is available in comicprocessor.py.
The metadata is read from the CBZ/ZIP comment as the ComicBookInfo format. Metadata from here does not contain Table of Content.
The script also read from ComicInfo.xml metadata. Here, the bookmarks are treated as table of contents, the same way as Kindle Comic Converter does. However, folder structure is not considered for the table of content.
Reading direction (LTR/RTL) is not stored nor read. This should be passed to the program as a parameter.
AZW3 is internally converted to EPUB using KindleUnpack. EPUB produced by this script is valid Kindle Comic EPUB.
PDF is currently read using ImageMagick to convert each page individually to image. This is a very slow process. Reading direction is read from the file, however, currently metadata and table of content is not read.
Table of content and reading is written to the output file. But non-latin may not display correctly in Adobe Reader.
This program is licensed under GNU General Public License 3.0 or later. The distribution include the KindleUnpack program available from https://github.com/kevinhendricks/KindleUnpack.