tanuki

README

Tanuki

Self-hosted manga server + reader

Features

• OPDS Support
• Multi-user support
• Supported formats: .cbz, .zip, .cbr, .rar
• Nested folders in library
• Track reading progress
• Thumbnail generation
• Single binary (~19.1 MB)
• Dark/light mode
• Metadata editor
• Mangadex downloader
• Responsive Desktop & Mobile UI
• Webtoon support (no row gaps)

Installation

Build from Source

# Clone the repo
$ git clone https://github.com/fiwippi/tanuki.git

# Change the working directory to tanuki
$ cd tanuki

# Build the app and run it
$ make build && make run

Docker

1. Clone the repository
2. Configure the docker-compose.yml file to set Tanuki to use the correct ports and mounted folders
3. Run docker-compose up
4. Open localhost:8096 or another port if you specified one

GitHub Container Registry

An official container image exists at ghcr.io/fiwippi/tanuki:latest

Usage

CLI

Usage of tanuki:
  -config string
        path to the config file, if it does not exist then it will be created (default "./config/config.yml")

Config

Tanuki runs using a config which has a default path of ./config/config.yml The config options and default values are specified below

---
host: 0.0.0.0
port: "8096"
logging:
  level: info
  log_to_file: true
  log_to_console: true
paths:
  db: ./data/tanuki.db
  log: ./data/tanuki.log
  library: ./library
session_secret: tanuki-secret
scan_interval_minutes: 180
max_uploaded_file_size_mib: 10
debug_mode: false

• To disable logging set logging.log_to_file and logging.log_to_console to false
• logging.level can be trace, debug, info, warn, error, fatal, panic
• scan_interval_minutes can be any non-negative integer. To disable interval scanning set the value to 0;

Initial Login

On the first run Tanuki logs the default username and randomly generated password to STDOUT. It is advised to immediately change the password

⚠️ - Tanuki generates all the thumbnails on startup which will cause a slight initial delay using it until thumbnail generation is done

Rar Archives

If you supply tanuki with Rar archives (.rar, .cbr), their unarchive time to retrieve a single page is about 2 seconds compared to only milliseconds for a Zip archives (.zip, .cbz) due to library constraints. For this reason consider converting all your files into Zip archives

OPDS

• The route for the OPDS catalog is /opds/v1.2/catalog
• This is the current OPDS feature support:

v1.2

• Basic Auth
• Viewing library
• Downloading archive
• Getting cover/thumbnail of archive
• Search
• Page streaming

Changelog

[0.16.2] - 2022-02-18

• Support Mangadex download with no chapter and volume

[0.16.1] - 2022-02-17

• Fixed bugs

[0.16] - 2022-02-15

• Tanuki automatically scans for new downloads once all downloads are done

[0.15] - 2022-02-15

⚠️ - Breaking changes are made to the database so you should recreate yours!

• Fixed bugs when tracking user progress
• Session token refreshed regardless of time left
• More informative errors and info added to the backend

[0.14] - 2022-02-07

• More responsive UI
• Removed unused internal packages
• License changed to BSD-3-Clause
• Default session cookie life changed to 3 days
• Updated to new Mangadex API change for downloading chapters
• Archive page sorting interprets underscores before other characters
• Remembers last page user tried to access if they're not authed anymore and redirects there after subsequent login

[0.13.1] - 2021-11-03

• Removed trailing dots from names of downloaded manga
• Default scan interval changed to 3 hours

[0.13] - 2021-11-03

• Stopped program consuming exhaustive amount of file descriptors
• Filepath sanitiser does not remove apostrophes and exclamation marks from downloaded Mangadex titles
• Autofocus for searchbar for Mangadex
• No autocorrect and autocapitalisation for the login screen
• Minor documentation changes

[0.12.3] - 2021-09-27

• Fixed page ordering bug introduced in 0.12.1

[0.12.2] - 2021-09-26

• More edge cases handled when decoding images

[0.12.1] - 2021-09-23

• Fixed ordering for some pages

[0.12] - 2021-09-23

• Better support for unicode characters
• Correct page displayed instead of page with similar name
• Loading animation for viewing the DB + searching and downloading from Mangadex
• Updated Mangadex functionality broken by Mangadex API change
• Series progress calculation takes into account all entries instead of only ones which have been accessed

[0.11] - 2021-08-24

• Thumbnail generation generated lazily instead of with a recurring job, this stops the throttling of the db whenever the job was running and speeds up page loading
• Mangas with pages specified only by numbers (with no text) no longer sorted in inverse order
• Stopped invalid progress displaying as NaN client-side
• Natural ordering improved so entries in a series now assigned correct order (until I manage to break it again)
• Invalid API requests automatically redirect the user to the login page as opposed to only when accessing the frontend routes
• When setting progress for a whole series, entry progress is created if it is previously nil
• Loading animation when scanning in the db or manually generating thumbnails

[0.10] - 2021-08-17

• Non-english titles displayed for Mangadex entries if english ones don't exist
• Mangadex archive pages zero padded in the archive
• Archives walked in proper lexical order (but lowercase first)
• Entry progress only created when reading the entry instead of when accessing the series
• Default entry progress is displayed as 0.00% instead of N/A
• Not all thumbnails have to be loaded to display them client side
• Series in the catalog and entries in each series are ordered naturally

[0.9.3] - 2021-08-13

• Stops CORS policy error when searching for Mangadex entries

[0.9.2] - 2021-08-13

• Removed delay on page load where navbar mobile shows until it detected it was on desktop

[0.9.1] - 2021-08-13

• If unspecified intervals in config then uses defaults

[0.9] - 2021-08-13

• Added favicon
• Added more documentation
• Docker image supports HTTP requests
• Docker image on the GitHub Container Registry
• Ignored correct files in .gitignore

[0.8] - 2021-08-12

• Specifying a zero/negative interval disables periodic tasks
• Ability to specify config path with param
• Stops thumbnail generation blocking populating the DB
• Added OPDS Search & Page Streaming

[0.7] - 2021-08-12

• Mangadex downloader

[0.6] - 2021-08-06

• Webtoon support, in the reader modal you can select the webtoon option which removes gaps between rows
• Mobile UI overhaul, tanuki is now smoother and works better on mobile

[0.5] - 2021-08-05

• Redundant CSS and JS code removed
• CSS classes all name appropriately
• CSS colours all specified in minify.go which makes it simpler to organise colours for both CSS themes
• Theme does not jitter on load anymore
• Alpine.js behaviour encapsulated into smaller classes which
  • enabled minification
  • reduces duplicity because behaviour is reproduced
• No more can't find img.src when loading images in the library view
• Many bug fixes, notable selecting options in the modal works correctly

[0.4] - 2021-07-30

• Assets minified before embedding
• Data preloaded through templating engine to reduce API calls

[0.3] - 2021-07-28

• Missing entries renamed to missing items
• Fixed bug where tanuki overwrites series/entry metadata with incorrect ones
• Tanuki now recognises missing series and entries as missing items
• Restructured project structure
  • Packages used by multiple main packages moved to the /internal directory
  • Packages used specifically to the user in the /pkg directory

[0.2] - 2021-07-23

• Go routines speed up parsing series and adding them to the DB
• Properties like user progress bundles into one API call so less calls are made
• Series list is stored in order as catalog instead of being generated on the fly
• Metadata for each entry is stored in order instead of being generated on the fly
• Parsing/adding/generating series/thumbnails doesn't stop at the first error, it returns all errors which occurred at the end

[0.1] - 2021-07-19

• Initial commit

License

BSD-3-Clause