How yukunsun/leek works
Project Type Analysis
This is an automated notification system that:
• Runs on GitHub Actions
• Scrapes financial data from external sources
• Sends notifications via webhook integrations
• Focuses on two main functions: fund valuation and bond subscription alerts
Main Components
a) External Data Sources (Input Layer):
• DanJuanApp (Fund valuations)
• TongHuaShun Data Center (Bond data)
b) GitHub Actions Layer:
• Two separate workflows (.github/workflows/)
• app-leek-bond.yml
• app-leek-fund.yml
c) Core Processing Components:
• Bond Module (bond/applyNewBond.py)
• Fund Module (fund/valuationTable.py)
• Utility Module (util/webHookUtil.py)
d) Notification Endpoints:
• Enterprise WeChat Bot
• Other webhook-compatible platforms (Feishu, etc.)
[External Data Sources] → [GitHub Actions Workflows] → [Processing Modules] → [Webhook Integration] → [Notification Platforms]
Specific Implementation Details
1. External scrapers collect data from financial sources
2. GitHub Actions triggers the workflows on schedule
3. Processing modules analyze and format the data
4. Webhook utility sends formatted messages
5. Notification platforms receive and display alerts
b) Configuration elements:
• GitHub Secrets (WEB_HOOK_BOND, WEB_HOOK_FUND)
• Requirements from requirements.txt
Important Architectural Notes
• Serverless nature of the system
• Separation of concerns between bond and fund modules
• Common webhook utility used by both workflows
• Configuration management through GitHub Secrets