CBZerv

A lightweight webserver for viewing mangas/ comics stored on a linux/ bsd server.

Features:

• Tag based search (with exclude support) (tagfile.txt)
• Directory preview pictures (folder.extension)
• No restriction on file-structure
• Cbz and Pdf support
• Phone-view support
• Low resource usage
• Hackable
  • Support for custom index.html files at any level (as well as accompanying css/.. files)
  • Usable with a reverse proxy
• Support for .ignore files
• Support for .ff.bz2 images within cbz (requires ImageMagick7 to be installed)

Usage:

Executable

PORT=8080 python3 cbzerv.py

Just execute it with python3.

Configuration (environmental variables):

• PORT: which port should be used (default: 8080)
• PWD (aka current working directory): the base directory

Filestructure

It can be structured however you want.

However: it is recommended to put each manga/comic into its own directory even if it has just one chapter.

Also: there are a few files with special meanings:

• tagfile.txt: a list of tags, which apply to a directory (manga/comic). One tag per line.
• folder.png, folder.jpg, etc: a picture representing the directory (example: cover, author-picture, etc)
• .ignore: do not include contents of this directory in listings or searches.
• index.html: override the listing of a directory with a custom one.

And comics/mangas should have a .cbz or .pdf file extension.

Example filestructure:

cbzerv.py
cbzerv
|- Marvel
|  |- folder.png
|  |- Spiderman
|     |- tagfile.txt
|     |- folder.jpg
|     |- Episode1.cbz
|     |- Episode2.pdf
|- Manga
   |- Izumi Tomoki      # <-- author
   |  |- Mieruko-chan
   |     |- tagfile.txt
   |     |- Ch 01.cbz
   |     |- Ch 02.cbz
   |- Tower of God
      |- Book 001.cbz

Example tagfile.txt contents:

Comedy
Slice of Life
Vampires
Shounen
Japanese

Scripts included

• downloader/guya.py: Synchronise all (or specific) mangas from Guya instances (should also support most guya-forks). (usage: python3 downloader/guya --help)
• tools/cbz_optimizer.nu: Try to reduce the cbz filesize without loosing data. (usage: nu tools/cbz_optimizer.nu --help)

Performance

Test environment:

• Raspberry Pi 4
• 32985 cbz files
• 1740 tagfiles

Resource Usage

Over 3 days including a lot of reading (measured using systemd):

• 49.921s CPU time
• RAM:
  • idle after 3 days: 10.0M + 11.1M swap
  • peak: 462.3M + 13.9M swap

Speed

Searching all files by tag (measured via firefox devtools):

• after wiping cache: about 370ms
• with intact cache: about 200ms