nyora 0.3.3

Pure Python implementation of the Nyora manga engine

Nyora — Python

Read like the world can wait.

The official Python package for Nyora — script your library, search 1000+ manga sources, and fetch chapters and pages straight from Python. Pure Python: no JVM, no desktop app, no Node.js, no Java. Just pip install and you're scripting manga in 60 seconds.

---

nyora is a pure-Python library, command-line tool, and terminal reader for the Nyora manga engine. Import it to drive 1000+ sources from your own code, pipe JSON into your scripts, or read manga in your terminal — installed with a single pip install, with no JVM, no Node.js, no Java, and no companion app to launch. It's fully open source, ad-free, and makes zero analytics or telemetry calls — the only network traffic is to the sources you ask for and the signed parser bundle.

pip3 install nyora

from nyora import Nyora

with Nyora() as client:
    source = client.sources.find("mangadex")
    page = client.manga.popular(source.id)
    print(page.entries[0].title)          # you're scripting manga. that's it.

📖 Full documentation: nyora.pages.dev/docs/python

---

Table of contents

• Why you'll love it
• 60-second quickstart
• About
• Python library (pip install nyora)
• Command line (nyora-cli)
• Installation
• What it can and cannot do
• FAQ
• Contributing
• Development setup
• Where things live
• Build from source
• Nyora on every platform
• Privacy & open source
• Acknowledgements
• License

---

Why you'll love it

• Installs in one line, runs anywhere Python runs. Pure Python with Python-installed dependencies — nothing to compile, no JVM, no Node.js, no Java, no desktop app.
• 1000+ sources, one API. Search, browse popular/latest, pull full chapter lists, and resolve page image URLs through a single clean, typed client.
• Sync and async. Drive it sequentially with Nyora, or fan out across many sources concurrently with AsyncNyora — same namespaces, same shapes.
• Read in your terminal. A full Textual-based TUI ships in the box: bare nyora-cli opens it. Pick a source, browse, open a title, page through a chapter — without leaving the shell.
• Scriptable by design. Add --json to any command and pipe straight into jq or your own tooling.
• Always current. Sources update over the air — sha256-verified, atomic — so new and fixed parsers arrive without upgrading the package.
• Private by default. No accounts, no ads, no tracking, no telemetry. Fully auditable, Apache-2.0 open source.

---

60-second quickstart

pip3 install nyora

Read in your terminal — zero code:

nyora-cli                              # launches the terminal reader (TUI)

Or script it — find a source, search, fetch pages:

from nyora import Nyora

with Nyora() as client:
    source = client.sources.find("asura")                 # fuzzy name or id
    results = client.manga.search(source.id, "Solo Leveling")
    entry = results.entries[0]

    details = client.manga.details(source.id, entry.url, title=entry.title)
    pages = client.manga.pages(source.id, details.chapters[0].url)

    for p in pages:
        print(p.url)                                       # page image URLs

Or stay in the shell and pipe JSON:

nyora-cli --json popular -s mangadex -p 1 | jq '.entries[].title'

That's the whole loop: install, pick a source, browse or search, fetch pages. Everything below goes deeper.

---

About

Nyora is a fast, free, ad-free, open-source manga reader that runs on every platform — with whole-page AI translation, 1000+ sources, offline downloads, and free cloud sync across all your devices. nyora brings that same source-and-parser engine to Python: a pip-installable library, a command-line tool (nyora-cli), and a terminal reader (TUI).

It runs the full Nyora parser bundle in-process through Python-installed dependencies — pure Python, end to end, with nothing to compile and no companion app to launch.

This single install gives you both surfaces below — a Python library you import, and the nyora-cli command-line tool. They are documented as clearly separate surfaces.

---

Python library (pip install nyora)

import nyora to drive Nyora's source/parser engine from your own code. Open a Nyora() client, find a source, search it, fetch details, and resolve page image URLs — all with a clean, typed API. No JVM helper, no desktop app, no Node.js, no Java.

from nyora import Nyora

with Nyora() as client:
    source = client.sources.find("mangadex")          # resolve by id or fuzzy name
    page = client.manga.popular(source.id)            # SearchPage of entries
    entry = page.entries[0]

    details = client.manga.details(source.id, entry.url, title=entry.title)
    pages = client.manga.pages(source.id, details.chapters[0].url)

    for p in pages:
        print(p.url)

The client exposes two typed namespaces:

• client.sources — list() the full bundled catalogue, or find(...) a source by id or fuzzy name (e.g. "asura").
• client.manga — popular(...), latest(...), search(...), details(...), and pages(...).

Browse popular and latest

from nyora import Nyora

with Nyora() as client:
    src = client.sources.find("mangadex").id

    popular = client.manga.popular(src, page=1)
    latest = client.manga.latest(src, page=1)

    for entry in popular.entries[:5]:
        print(entry.title, "—", entry.url)

Async fan-out across many sources

For concurrent fan-out across many sources, use the async client, which exposes the same namespaces:

import asyncio
from nyora import AsyncNyora

async def main():
    async with AsyncNyora() as client:
        source = await client.sources.find("asura")
        results = await client.manga.search(source.id, "Solo Leveling")
        print(results.entries[0].title)

asyncio.run(main())

Attach to a running helper

You can also attach to an already-running Nyora helper (the desktop helper, another app, or nyora-cli serve) over its REST contract instead of running the engine in-process:

from nyora import NyoraHelper

with NyoraHelper.attach("http://127.0.0.1:54123") as client:
    print(client.health())

Over-the-air source updates

The parser bundle and source catalogue update over the air, so new and fixed sources arrive without upgrading the package:

from nyora import Nyora

with Nyora() as client:
    available, installed, latest = client.check_update()
    if available:
        result = client.update()        # sha256-verified, atomic
        print("updated to OTA version", result.version)

→ Library guide: nyora.pages.dev/docs/python/guide/library · API reference: /reference/api

---

Command line (nyora-cli)

Installing the package adds the nyora-cli command (also aliased as nyora). It drives the same pure-Python engine as the library.

Running bare nyora-cli with no subcommand launches the terminal reader (TUI). Pass a subcommand to run a one-shot command instead.

nyora-cli                                    # no subcommand -> launches the TUI
nyora-cli --help                             # list all commands and options
nyora-cli sources --search asura             # one-shot subcommand
nyora-cli search -s asura "Solo Leveling"

Subcommands

Command	Description
sources [--search Q]	List the source catalogue, or fuzzy-find sources by name
search -s SRC [-p PAGE] QUERY	Search a source for a query
popular -s SRC [-p PAGE]	Browse a source's popular titles
latest -s SRC [-p PAGE]	Browse a source's latest updates
details -s SRC URL	Fetch manga details and the full chapter list
pages -s SRC CHAPTER_URL [--branch B]	Resolve page image URLs for a chapter
download -s SRC CHAPTER_URL [-o OUT]	Download a chapter as a single .cbz archive (OUT is a .cbz file or a directory; default <chapter>.cbz)
batch -s SRC MANGA_URL [-o DIR]	Batch download ALL chapters of a manga, one .cbz per chapter, into DIR
update [--force]	Self-update the parser bundle over the air (OTA)
serve [--host H] [--port P]	Run the pure-Python REST helper
version	Print the package and installed OTA version

-s/--source accepts a source id or a fuzzy name (e.g. asura). Add the global --json flag to any subcommand to emit raw JSON instead of a pretty table — ideal for piping into jq or wiring into scripts:

nyora-cli --json popular -s mangadex -p 1

Terminal reader (TUI)

Run bare nyora-cli (or nyora-tui) to open the full terminal reader, built on Textual: pick a source, browse popular/latest/search, open a title, and page through a chapter — all without leaving the shell. It is non-TTY safe: in a non-interactive shell it prints a friendly notice and exits cleanly.

nyora-cli        # launches the TUI
nyora-tui        # also launches the TUI

Pure-Python REST helper

nyora-cli serve starts a small stdlib HTTP server that exposes the engine over the same camelCase REST contract the desktop helper uses, so any other Nyora app (or NyoraHelper.attach(...)) can connect:

nyora-cli serve --host 127.0.0.1 --port 0
# -> http://127.0.0.1:54123

It binds the requested host/port (port 0 picks a free port), prints the base URL, and writes a helper.port file so other Nyora processes can auto-discover it. Endpoints include /health, /sources, /sources/popular, /sources/latest, /sources/search, /manga/details, and /manga/pages.

→ CLI guide: /guide/cli · TUI guide: /guide/tui · Server guide: /guide/server

---

Installation

From PyPI (recommended)

pip3 install nyora

This installs the nyora Python library and the nyora-cli command (aliased as nyora), including the Textual-based terminal reader and the REST helper. It is pure Python with only Python-installed dependencies — no JVM helper, desktop app, Node.js, or Java to install.

Frictionless and safe by design. nyora is published to PyPI straight from this public, Apache-2.0 repository — you can read every line before you run it. There's no account to create, no ads, no tracking, and no telemetry. The only network calls it makes are to the sources you explicitly ask for and to fetch the sha256-verified OTA parser bundle. Your library, history, and downloads stay on your machine, under your control.

Install	Command	Adds
Default	pip3 install nyora	Library + nyora-cli + TUI + REST helper
Docs tooling	pip3 install "nyora[docs]"	Sphinx + Furo to build these docs
Dev tooling	pip3 install "nyora[dev]"	Build, test, lint, and publish tooling

Tip — keep it isolated. If you'd rather not touch your global environment, install into a virtual environment (python -m venv .venv && source .venv/bin/activate) or use pipx (pipx install nyora) to get the CLI on your PATH in its own sandbox.

Requirements

• Python 3.10 or newer (tested through 3.14).
• A network connection for source requests and OTA parser-bundle updates.

Updating

Two layers update independently:

• The package (library + CLI): pip3 install --upgrade nyora.
• The sources (parser bundle + catalogue): nyora-cli update (add --force to re-pull). New and fixed sources arrive over the air without a package upgrade.

Check what you're running with nyora-cli version.

Troubleshooting

• nyora-cli: command not found — ensure your Python Scripts/bin directory is on your PATH, or invoke it as python -m nyora.
• Stale or missing sources — run nyora-cli update (or --force), then nyora-cli version to confirm the installed OTA version.
• Permission errors writing the cache — the OTA bundle is written into the user cache directory; make sure that location is writable for your user.

---

What it can and cannot do

Capability	Supported	Notes
List the full source catalogue	Yes	client.sources.list() / nyora-cli sources
Resolve a source by id or fuzzy name	Yes	client.sources.find(...)
Popular / latest / search browsing	Yes	client.manga.popular · .latest · .search
Manga details + full chapter list	Yes	client.manga.details(...)
Resolve page image URLs	Yes	client.manga.pages(...)
Download a chapter to a .cbz archive	Yes	nyora-cli download
Synchronous and async clients	Yes	Nyora and AsyncNyora
Run as a REST helper / attach to one	Yes	nyora-cli serve · NyoraHelper.attach(...)
OTA self-update of sources	Yes	sha256-verified, atomic writes
Pure Python — no JVM / Node.js / Java	Yes	Runs the parser bundle in-process
Host the consumer reading UI	No	Use the platform apps for a full reader
Bundled OCR / image translation pipeline	No	Translation lives in the consumer apps; the library gives you the page URLs to build on
Bypass a source's own access controls	No	It parses publicly accessible providers only

---

FAQ

Is it really free? Yes. nyora is 100% free and open source under Apache-2.0. There are no paid tiers, no ads, and no upsells.

Do I need an account? No. There's nothing to sign up for. Install the package and start reading or scripting immediately.

Will my data be private? Yes. There are no analytics, no telemetry, and no accounts. The only network calls nyora makes are to the sources you explicitly request and to fetch the sha256-verified OTA parser bundle. Your library, history, and downloads live on your own machine.

Is it safe? Can I audit it? Every line is public, Apache-2.0 source code in this repository, and the wheel on PyPI is built from it. You're welcome to read the code, run it in a virtual environment, and inspect exactly what it talks to over the network.

How do I update? Upgrade the package with pip3 install --upgrade nyora, and refresh the sources with nyora-cli update. The two update independently — most source fixes arrive over the air without a package upgrade.

Do I have to write code to use it? No. Run bare nyora-cli to open the terminal reader (TUI) and browse with the keyboard, or use one-shot subcommands like nyora-cli search -s asura "Solo Leveling". The Python API is there when you want it.

A source stopped working — what do I do? Run nyora-cli update --force first; many breakages are fixed in the OTA bundle. If it persists, please open an issue so it can be tracked and fixed.

Does it translate manga like the apps do? Not in the box. Whole-page OCR/translation lives in the consumer apps. nyora gives you the page image URLs so you can build that (or anything else) on top.

---

Contributing

Contributions are genuinely welcome — and you can start today. nyora is fully open Python: there's no private engine, no closed core, and no special access required. If you can run uv sync, you can run the whole thing and change anything you like.

Ways to contribute (every skill level)

You don't need to be a Python expert — or even write code — to help:

• Report a bug. Hit a broken source, a confusing error, or a crash? Open an issue with what you ran and what happened. Clear bug reports are worth a lot.
• Request or help a source. Want a source supported, or noticed one parsing wrong? File an issue. Sources are driven by the OTA parser bundle, so flagging breakage helps everyone.
• Improve the docs. The Sphinx docs live in docs/ (guide/ and reference/). Typos, clearer examples, and new how-to recipes are all great first PRs.
• Test releases. Try a new version on your platform and Python (3.10–3.14) and report what works or breaks — especially the CLI and TUI in different terminals.
• Improve or translate the UI strings. The Textual TUI lives in src/nyora_tui/; clearer wording and better keyboard flows are welcome.
• Write code. Add API examples, sharpen typing, fix bugs, or extend the CLI/TUI. See "Good first contributions" below.
• Star and share. Genuinely one of the most helpful things you can do — it helps other readers find the project. If nyora saved you time, a star and a link go a long way.

Good first contributions

Pulled from how this repo is actually laid out:

• A new CLI subcommand. Subcommands live in src/nyora/cli.py and call into the client.sources / client.manga namespaces — a small, self-contained surface to add to. Mirror an existing one and add a row to the subcommands table above.
• A new library example or recipe. Add a focused snippet to docs/guide/library.md showing a real task (e.g. exporting a chapter list to CSV, fanning out a global search with AsyncNyora).
• A TUI nicety. src/nyora_tui/app.py is a single Textual app — small ergonomic fixes (key bindings, empty-state messages, status text) are friendly first changes.
• A REST endpoint or --json polish. The stdlib server in src/nyora/server.py and the --json output in cli.py are easy places to add value for scripters.
• Docs and typing. Tightening docstrings, type hints, or the API reference in docs/reference/api.md is always appreciated.

PR & issue etiquette

• Keep PRs focused. One change per PR is much easier to review and merge than a grab-bag.
• Describe the change. A sentence or two on what and why, plus how you tested it, is plenty.
• Run the checks first (see Development setup): ruff, mypy, and pytest should pass.
• Be kind. Reviews are a conversation. Newcomers are welcome and questions are fine — ask early rather than guess.
• Match the existing style. Typed, dependency-light, pure Python. If you're unsure where something belongs, open an issue first and we'll point you at the right file.

Issues and discussion happen here: github.com/hasan72341/Nyora/issues.

---

Development setup

This is the contributor quickstart (distinct from the end-user pip install). The project is managed with uv.

# 1. Clone
git clone https://github.com/Hasan72341/nyora-python
cd nyora-python

# 2. Install everything (deps + dev + docs tooling) into a managed venv
uv sync --extra dev --extra docs

# 3. Smoke test — confirm the core symbols import cleanly
uv run python -c "from nyora import Nyora, Manga, Source; print(Nyora, Manga, Source)"

# 4. Run it
uv run nyora-cli sources --search asura
uv run nyora-cli                       # launches the TUI

Prerequisites: Python 3.10+ and uv. That's it — pure Python, nothing to compile, no JVM/Node.js/Java.

Run the checks (please do this before opening a PR):

uv run ruff check .          # lint
uv run ruff format .         # format
uv run mypy                  # type-check (packages: nyora, nyora_tui)
uv run pytest                # tests (live in tests/)

Where to look first: start in src/nyora/client.py to see how the sources and manga namespaces are wired, then follow into src/nyora/services/ for the actual operations. For the CLI, read src/nyora/cli.py; for the TUI, src/nyora_tui/app.py.

Bumping the bundled sources (internal)

The parser bundle (parsers.bundle.js) and source catalogue (sources.json) are force-included into the wheel (see pyproject.toml) and otherwise refreshed over the air at runtime. Day-to-day contributors don't need to touch these — source fixes ship through the OTA channel, not the package.

---

Where things live

A quick map so you can navigate the repo:

Path	What's there
src/nyora/client.py	The Nyora / AsyncNyora clients and the sources / manga namespaces
src/nyora/services/	The operations behind the namespaces — sources.py, manga.py, downloads.py, library.py, backup.py, system.py
src/nyora/models.py	Typed dataclasses (Source, Manga, MangaDetails, MangaChapter, SearchPage, …)
src/nyora/cli.py	The nyora-cli command, subcommands, and --json output
src/nyora/server.py	The stdlib REST helper (nyora-cli serve)
src/nyora/ota.py	Over-the-air bundle updates (sha256-verified, atomic)
src/nyora/runtime.py, parser_bridge.py, direct.py	The in-process engine and parser bridge
src/nyora/helper.py	NyoraHelper.attach(...) for talking to a running helper
src/nyora_tui/app.py	The Textual terminal reader
tests/	The pytest suite (test_cli.py, test_server.py, test_ota.py, …)
docs/	Sphinx docs — guide/ and reference/
pyproject.toml	Project metadata, dependencies, and tool config (ruff, mypy, pytest)

---

Build from source

For local development in this repository, the project is managed with uv.

uv sync --extra dev --extra docs
uv run python -c "from nyora import Nyora, Manga, Source; print(Nyora, Manga, Source)"

uv sync resolves and installs all dependencies into a managed virtual environment, and the uv run smoke test confirms the core symbols import cleanly. Requires Python 3.10+.

Build the documentation locally with:

bash scripts/build-docs.sh

Packaging

uv lock
uv build
uv run twine check dist/*

The parser bundle and source catalogue are force-included into the wheel, so a fresh install can run the engine immediately and update over the air from there.

---

Nyora on every platform

The Nyora reader is everywhere your screens are — and your library, history, bookmarks, and progress sync for free across all of them.

Platform	Repo	Get it
Python	nyora-python (you are here)	pip3 install nyora
Android	nyora-android	APK
macOS	nyora-mac	.dmg / brew
Windows	nyora-windows	.exe (x64/ARM64)
Linux	nyora-linux	deb · rpm · curl
iOS / iPadOS	nyora-ios	sideload IPA
Web	nyora-web	nyoraweb.pages.dev

Want a big, mechanical, high-impact contribution? The iOS engine (NyoraEngine) is porting roughly 1,300 sources as mostly-mechanical template subclasses — the framework and one template are done, leaving ~1,331 parsers to port across 3,659 classes. It's highly parallelizable and the headliner "help wanted" across the project. If that's your kind of thing, it's a great place to make a dent.

---

Privacy & open source

Nyora is 100% free, ad-free, and contains no tracking. nyora is fully auditable open-source code: there are no analytics, no telemetry, and no accounts. The only network calls it makes are to the sources you ask for and to fetch the sha256-verified OTA parser bundle. Licensed under Apache-2.0.

Acknowledgements

Nyora's source and parser engine builds on the work of the open-source manga community. nyora is developed and maintained by Md Hasan Raza — GitHub · [e-mail].

If nyora is useful to you, the kindest thing you can do is star the repo and share it — it helps other readers and scripters find the project, and it's the simplest way to say thanks. Pull requests, however small, are always welcome.

License

Licensed under Apache-2.0. See the project metadata in pyproject.toml for details.

---

Nyora is not affiliated with any of the manga sources it can access.