Skip to content

Replace Processing Pipeline table with Mermaid flowchart#3

Closed
rafacm wants to merge 1 commit intomainfrom
readme-mermaid-pipeline
Closed

Replace Processing Pipeline table with Mermaid flowchart#3
rafacm wants to merge 1 commit intomainfrom
readme-mermaid-pipeline

Conversation

@rafacm
Copy link
Copy Markdown
Owner

@rafacm rafacm commented Mar 10, 2026

Summary

  • Replaced the markdown table in the Processing Pipeline section with a Mermaid flowchart diagram
  • Shows the 12-step pipeline as a vertical flow with failure edges to a "Failed" node
  • Color-coded: green for "Ready", red for "Failed"

Test plan

  • Verify the Mermaid diagram renders correctly on GitHub

🤖 Generated with Claude Code

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
@rafacm
Copy link
Copy Markdown
Owner Author

rafacm commented Mar 10, 2026

Closing since the Mermaid diagram is too complex

@rafacm rafacm closed this Mar 10, 2026
@rafacm rafacm deleted the readme-mermaid-pipeline branch March 21, 2026 04:45
rafacm added a commit that referenced this pull request Mar 31, 2026
…protection

- Remove wikidata_id entirely from resolver schema/prompt — linking
  agent is the single owner of Q-ID assignment (comment #1)
- Pass entity_type_key filter to run_linking_agent from management
  command (comment #2)
- Show remaining pending count after batch completes (comment #3)
- Add LINKING status for atomic work claiming to prevent concurrent
  tasks from processing the same entities (comment #4)
- Clarify admin retry action messaging (comment #5)
- Narrow except to ImportError in apps.py (comment #6)
- Add explanation for unavailable session IDs (comments #7, #8)
- Add RAGTIME_LINKING_AGENT_* settings to settings.py so .env values
  are picked up (comment #9)

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
rafacm added a commit that referenced this pull request Apr 13, 2026
1. Add start_from param to run_pipeline/route_entry so admin reprocess
   overrides data-based routing instead of silently skipping steps
2. Add _close_connections() for DB connection management in background
   threads (gated on non-main thread to avoid breaking tests)
3. Route ALL step failures to recovery node (not just scraping/downloading)
   so RecoveryAttempt records and human escalation work for every step
4. Doc now accurate after fix #3
5. Preserve original error_type/http_status/exception_class from
   PipelineEvent using dataclasses.replace() in handle_step_failure_from_graph
6. Fix telemetry.py module docstring re: OTel import when disabled
7. embed_node() fails fast instead of marking READY without embeddings
8. Add test_graph.py with 12 tests for route_entry, after_step,
   after_recovery, and compiled graph integration
9. Add comment to apps.py explaining signals kept as extension points
10. Fix _build_exporter() docstring

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant