Self-hostable progress tracking manga reader
scheduler: add latest chapter number to log
log search query in mangadex error msg
add .envrc



You can also use your local clone with git send-email.

Live demo: https://pytaku.imnhan.com (db may be hosed any time, also expect bugs)

Production instance coming When It's Ready (tm).

#Pytaku builds.sr.ht status

Pytaku is a WIP web-based manga reader that keeps track of your reading progress and new chapter updates. Its design goals are:

  • Self-host friendly - if you have a UNIX-like server with python3.7+ and can run pip install, you're good.

  • Phone/tablet friendly - although I hardly read any webtoons these days so the phone experience may not be as polished.

  • KISSFFS, or Keep It reaSonably Simple you F-ing architecture/tooling FetishiSts! Oftentimes I have enough practice on industrial grade power tools at work so at home I want a change of pace. Flask + raw SQL has been surprisingly comfy. On the other side, mithril.js provides a good baseline of SPA functionality without having to pull in the Rube Goldberg machine that is """modern""" JS devtools.

#Keyboard shortcuts

On Chapter page, press ? to show keyboard shortcuts.


## Backend ##
poetry install

pytaku-generate-config > pytaku.conf.json
# fill stuff as needed

# run migration script once

# run 2 processes:
pytaku-dev -p 8000  # development webserver
pytaku-scheduler  # scheduled tasks e.g. update titles

## Frontend ##

sudo pacman -S entr  # to watch source files
npm install -g --prefix ~/.node_modules esbuild # to bundle js

# Listen for changes in js-src dir, automatically build minified bundle:
find src/pytaku/js-src -name '*.js' | entr -rc \
     esbuild src/pytaku/js-src/main.js \
     --bundle --sourcemap --minify \

#Dumb proxy

Eventually mangasee started using a somewhat aggressive cloudflare protection so cloudscraper alone is not enough (looks like our IP got blacklisted or throttled all the time), so now I have to send requests through a crappy GAE-based proxy. You'll need to spin up your own proxy instance (Google App Engine free tier is enough for personal use), then fill out OUTGOING_PROXY_NETLOC and OUTGOING_PROXY_KEY accordingly.

Yes it's not a standards-compliant http(s) proxy so you can't just use yours. I chose the cheapest (free) way to get a somewhat reliable IP-rotating proxy.


Can be run with just pytest. It needs a pytaku.conf.json as well.

#Code QA tools

  • Python: black, isort, flake8 without mccabe
  • JavaScript: jshint, prettier
sudo pacman python-black python-isort flake8 prettier
npm install -g --prefix ~/.node_modules jshint


Gotcha: mangasee image servers will timeout if you try to download images via ipv6, so you'll need to disable IPv6 on your VM. It's unfortunate that python-requests [doesn't][https://github.com/psf/requests/issues/1691] have an official way to specify ipv4/ipv6 on its API, and I'm too lazy to figure out alternatives.

I'm running my instance on Debian 11, but any unix-like environment with these should work:

  • python3.7+
  • the rest are all pypi packages that should be automatically installed when you run pip install pytaku

The following is a step-by-step guide on Debian 11.

sudo apt install python3-pip
pip3 install --user pytaku
# now make sure ~/.local/bin is in your $PATH so pytaku commands are usable

pytaku-generate-config > pytaku.conf.json
# fill stuff as needed

# run migration script once

# run 2 processes:
pytaku -w 7  # production web server - args are passed as-is to gunicorn
pytaku-scheduler  # scheduled tasks e.g. update titles

# don't forget to setup your proxy, same as in development:
# https://git.sr.ht/~nhanb/gae-proxy

# upgrades:
pip3 install --user --upgrade pytaku
# then restart `pytaku` & `pytaku-scheduler` processes

If you're exposing your instance to the internet, I don't have to remind you to properly set up a firewall and a TLS-terminating reverse proxy e.g. nginx/caddy, right?

Alternatively, just setup a personal tailscale network and let them worry about access control and end-to-end encryption for you.

#Optional optimization

With the setup above, you're serving static assets using gunicorn, which is not ideal performance-wise. For private usage this doesn't really matter. However, if you want to properly serve static assets using nginx and the like, you can copy all static assets into a designated directory with:

pytaku-collect-static target_dir

This will copy all assets into target_dir/static. You can now instruct nginx/caddy/etc. to serve this dir on /static/* paths. There's an example caddyfile to do this in the ./contrib/ dir.


Copyright (C) 2021 Bùi Thành Nhân

This program is free software: you can redistribute it and/or modify it under the terms of the GNU Affero General Public License version 3 as published by the Free Software Foundation.

This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU Affero General Public License for more details.

You should have received a copy of the GNU Affero General Public License along with this program. If not, see https://www.gnu.org/licenses/.