pysaka 0.4.0

Async Python client for Sakamichi Group Message API (Nogizaka46, Sakurazaka46, Hinatazaka46)

pysaka

Disclaimer & Warnings

[!CAUTION] Use at your own risk. accessing the API via unauthorized means may violate the Terms of Service of the respective platforms. This library is for educational purposes only.

規約 / Terms of Service Ref.

Users must agree to the official Terms of Service of the respective platforms. The following are excerpts from the official terms (as of writing):

• Nogizaka46 Message App Terms
• Sakurazaka46 Message App Terms
• Hinatazaka46 Message App Terms

第3条（知的財産権）/ Article 3 (Intellectual Property)

3. 当社が別に定める場合を除き、お客様が本コンテンツを複製、翻案、頒布、公衆送信等することは禁止します。

第8条（禁止事項）/ Article 8 (Prohibited Acts)

(11) 当社または第三者の情報、データおよびソフトウェアを修正、改変、改ざん、リバースエンジニアリング、逆コンパイル、逆アッセンブルまたは消去等する行為

(16) 当社が指定するアクセス方法以外の手段で本サービスにアクセスし、またはアクセスを試みる行為

(17) 自動化された手段（クローラおよび類似の技術を含む）を用いて本サービスにアクセスし、またはアクセスを試みる行為

Async Python client for Sakamichi Groups (Nogizaka46, Sakurazaka46, Hinatazaka46, Yodel) Message API.

pysaka provides a robust, type-hinted, and async interface to interact with the official Message apps for all supported groups. It handles authentication (via browser), token management, and data retrieval.

Features

• 🔐 Browser Authentication: Seamless interactive login via Playwright (compatible with MFA/SSO).
• 🍪 Auto-Refresh: Automatically refreshes access tokens using captured cookies.
• 🚀 Async/Await: Built on aiohttp for high-performance concurrent requests.
• 📦 Multi-Group: Supports Nogizaka46, Sakurazaka46, Hinatazaka46, and Yodel out of the box.
• 📝 Blog Scraper: Backup official blogs (HTML + images) for all three groups.
• 🛠️ Type Hinted: 100% type coverage for better IDE support.

Configuration

pysaka uses structlog for observability. You can control the logging output via environment variables:

• HAKO_ENV=development (default): Pretty-printed, colored console logs.
• HAKO_ENV=production: Structured JSON logs with automatic secret redaction.

Installation

Recommended install via uv:

uv add pysaka

For development:

git clone https://github.com/xebjhm/pysaka.git
cd pysaka
uv sync

Quick Start

1. Authentication

Use BrowserAuth to log in interactively. This launches a browser window for you to enter credentials.

import asyncio
from pysaka import BrowserAuth, Group

async def login():
    creds = await BrowserAuth.login(Group.NOGIZAKA46)
    print(creds['access_token'])

asyncio.run(login())

2. Fetching Data

Initialize the Client with your credentials.

import asyncio
import aiohttp
from pysaka import Client, Group

async def main():
    # ... assume creds obtained via BrowserAuth ...
    token = "YOUR_ACCESS_TOKEN"

    async with aiohttp.ClientSession() as session:
        client = Client(Group.NOGIZAKA46, access_token=token)

        # Get Profile
        profile = await client.get_profile(session)
        print(f"Hello, {profile['nickname']}!")

        # Get Groups (Members)
        groups = await client.get_groups(session)
        for g in groups:
            print(f"{g['name']} (ID: {g['id']})")

asyncio.run(main())

3. Blog Scraping (No Auth Required)

Scrape official blogs for any group without authentication.

import asyncio
import aiohttp
from pysaka.blog import NogizakaBlogScraper, HinatazakaBlogScraper, SakurazakaBlogScraper

async def scrape_blogs():
    async with aiohttp.ClientSession() as session:
        scraper = NogizakaBlogScraper(session)

        # Get all active members
        members = await scraper.get_members()
        print(f"Found {len(members)} members")

        # Get blogs for a specific member
        async for blog in scraper.get_blogs(member_id="some_member_code"):
            print(f"{blog.title} - {blog.published_at}")

asyncio.run(scrape_blogs())

API Reference

Client

The main entry point.

• __init__(group, access_token, ...): Initialize client.
• get_profile(session): Get current user profile.
• get_groups(session): List subscribed members/groups.
• get_messages(session, group_id, ...): Fetch messages timeline.
• get_news(session): Fetch official announcements.
• get_tags(session): Fetch tags.
• get_fc_contents(session): Fetch Fan Club content.
• get_organizations(session): Fetch organizations.
• get_products(session, product_type): Fetch products (subscriptions).
• post_json(session, endpoint, data): Perform JSON POST requests.
• delete_json(session, endpoint): Perform DELETE requests.
• download_file(session, url, filepath): Download a file to local filesystem.
• get_member(session, member_id): Fetch individual member details.
• get_account(session): Fetch user account information.
• get_letters(session, group_id): Fetch user's sent letters/cards.
• get_past_messages(session, group_id): Fetch historical messages.
• get_subscription_streak(session, group_id): Fetch consecutive subscription days.
• add_favorite(session, message_id): Add a message to favorites.
• remove_favorite(session, message_id): Remove a message from favorites.
• refresh_if_needed(session): Lazy token refresh if expiring soon.

Credential Management

• get_token_manager(): Get the singleton TokenManager instance for secure credential storage.
• TokenManager.save_session(group, access_token, ...): Save session credentials to system keyring.
• TokenManager.load_session(group): Load stored session credentials.
• TokenManager.delete_session(group): Remove stored credentials.

BrowserAuth

Helper for OAuth2 flow.

• login(group, headless=False, ...): Perform login and capture tokens.

Blog Scrapers

Public blog scrapers (no authentication required).

• NogizakaBlogScraper(session): Nogizaka46 official blog.
• SakurazakaBlogScraper(session): Sakurazaka46 official blog.
• HinatazakaBlogScraper(session): Hinatazaka46 official blog.

Each scraper provides:

• get_members(): Get dict of member_id -> member_name.
• get_blogs(member_id, since_date=None): AsyncIterator of BlogEntry objects.
• get_blog_detail(blog_id, member_id=None): Fetch single blog by ID.

Exceptions

• SakaError: Base exception for all pysaka errors.
• AuthError: Authentication related errors.
• ApiError: API request errors (includes status_code attribute).
• SessionExpiredError: Session invalidated server-side (e.g., logged in from another device).
• RefreshFailedError: All token refresh attempts failed unexpectedly.
• BlogGoneError: Blog post has been permanently removed (HTTP 404/410).

Contributing

Contributions are welcome! Please read CONTRIBUTING.md for details.

License

MIT