How nanos/fedifetcher works
Main Components
a) Client Components:
• Multiple execution methods (represent as parallel entry points):
• GitHub Actions runner
• Cron job
• Container runtime
• Windows Task Scheduler
b) Core Components:
• Main Script Modules:
• find_posts.py (Primary execution module)
• get_context.py (Context fetching module)
c) External Systems:
• Mastodon API (Primary integration point)
• Other Fediverse Platforms:
• Pleroma/Akkoma
• Misskey/Calckey/Firefish
• Lemmy
• Peertube
d) Storage:
• Artifacts storage (for caching fetched posts)
[Client Layer]
• ↓
[Core Processing Layer]
• ↓
[External Integration Layer]
• ↓
[Storage Layer]
Component Details
For each component, show:
• Input/output relationships
• Key functionalities:
• Timeline fetching
• Reply fetching
• Profile backfilling
• Cache management
Technical Details
A technical metadata section showing:
• Python runtime requirement
• Docker container support
• API version compatibility
• Required permissions/scopes
The configurable nature of the fetching operations
Keep the abstraction level focused on the system architecture while including enough detail to show the unique aspects of FediFetcher, particularly its various fetching capabilities and caching mechanism.
Remember to indicate the asynchronous nature of the operations and the potential for long-running processes, as mentioned in the README's warning about execution time.