Nazurin

English | 中文

小さな小さな賢将, a Telegram bot which helps you collect ACG illustrations from various sites.

Architecture

Features

View/Download artwork from various sites
Add images to your collection via Telegram
Store your collection in Telegram channels
Store images on local disk or MEGA
Store image metadata in multiple types of database

Supported Sites

Name	URL	Commands	Collection
Pixiv	https://www.pixiv.net/	✔	✔
Danbooru	https://danbooru.donmai.us/	✔	✔
Safebooru	https://safebooru.donmai.us/		✔
yandere	https://yande.re/	✔	✔
Konachan	https://konachan.com/	✔	✔
Lolibooru	https://lolibooru.moe/		✔
Zerochan	https://www.zerochan.net/	✔	✔
Gelbooru	https://gelbooru.com/		✔
Twitter	https://twitter.com/		✔
Bilibili Dynamics	https://t.bilibili.com/		✔

Deploy

Deploy on Heroku

'Deploy to Heroku' Button:

Tips: You can fork this repository, modify it to your needs, and use this button to deploy your own version.

Manual deploy:

Set all required environment variables on Heroku according to config.py(root directory & /sites), clone this repository and push to Heroku, everything should be working properly.

Deploy on your own server

Tips: You may comment out unused dependencies in requirements.txt before installation.

Install dependencies: pip install -r requirements.txt
Set the required environment variables
Start the bot: python bot.py

Usage

Commands:

/ping - pong
/pixiv <id> - view pixiv artwork
/pixiv_download <id> - download pixiv artwork
/danbooru <id> - view danbooru post
/danbooru_download <id> - download danbooru post
/yandere <id> - view yandere post
/yandere_download <id> - download yandere post
/konachan <id> - view konachan post
/konachan_download <id> - download konachan post
/zerochan <id> - view zerochan post
/zerochan_download <id> - download zerochan post
/bookmark <id> - bookmark pixiv artwork
/clear_downloads - clear download cache
/help - get help text

How to update your collection

Send the bot a message with a link of supported sites, this message will be forwarded to GALLERY channel, the bot will then download the original images from the site, send the files to ALBUM channel, and finally store to your custom destination.

Tips: On mobile you can use the share button in apps, as long as the final message contains a link.

Configuration

ENV

The default option (production) uses Webhook mode, you can set to development to use polling mode.

TOKEN

API token of your bot, can be obtained from @BotFather.

WEBHOOK_URL

Webhook URL to be set for Telegram server.

PORT

Webhook port, automatically set if on Heroku.

STORAGE

String, evaluated to list.

Type of storage, default is [] which only uses DOWNLOAD_DIR as local storage. Set to ['Mega'] to use MEGA.

Implement other storage by creating a file under storage with a store function.

DOWNLOAD_DIR

Local directory to store downloaded images, will be created if not exists.

STORAGE_DIR

Storage directory, can be local or remote, will be created if not exists.

ALBUM_ID

Telegram channel ID used for storing files.

GALLERY_ID

Telegram channel ID used for storing messages, messages sent to bot and containing URL entities will be forwarded here for reviewing.

ADMIN_ID

Telegram user ID(not username) of the admin, bot functions are restricted to admin user.

Tips:

You can get your User ID & Channel ID via @GetIDs Bot

You need to add your bot to channel administrators

DATABASE

Type of database.

Supported databases:

Driver	URL	Config	Note
TinyDB	https://tinydb.readthedocs.io/en/stable/index.html	`Local`	Default
Firestore	https://firebase.google.com/products/firestore	`Firebase`
MongoDB	https://www.mongodb.com/	`Mongo`	MongoDB Atlas supported
Cloudant	https://www.ibm.com/cloud/cloudant	`Cloudant`

You can also implement your own database driver by creating a file under database folder, and set this option to the name of driver class.

GOOGLE_APPLICATION_CREDENTIALS

Firebase SDK credentials, see Firebase Documentation.

For Heroku, you can copy the content of service-account-file.json.

MONGO_URI

MongoDB connection string, must specify database.

eg: mongodb://username:password@localhost:27017/database, default is mongodb://localhost:27017/nazurin.

CLOUDANT_USER & CLOUDANT_APIKEY & CLOUDANT_DB

Cloudant username, API key and database name, using IAM authentication, default database is nazurin.

PIXIV_USER & PIXIV_PASS

Pixiv email or user id, and password.

MEGA_USER & MEGA_PASS

MEGA login email and password.

MEGA Encoding Issue

Due to unknown reasons, there're encoding issues with special filenames (e.g.:special or full-width characters) on MEGA. These filenames may be wrongly displayed on MEGA Android & iOS clients, and cannot be synced through MEGASync. However, MEGA Web client can handle these files correctly, so to solve this problem, do the following:

Create a temporary directory and set STORAGE_DIR to it
Periodically log into MEGA Web client
Copy all files from temporary directory to archive directory (Do not use move)
Delete files in temporary directory
Problem solved

Of course you can modify filename format to avoid this problem

SagiriShiho / nazurin