Skip to content

Harden distributed deduplication and adaptive execution#190

Merged
luigilcsilva merged 3 commits into
mainfrom
inconsitency-correction
Jul 6, 2026
Merged

Harden distributed deduplication and adaptive execution#190
luigilcsilva merged 3 commits into
mainfrom
inconsitency-correction

Conversation

@luigilcsilva

Copy link
Copy Markdown
Collaborator
  • keep Parquet and CSV consolidation distributed until final assembly
  • make draw_one deterministic and lazy for large catalogs
  • validate and normalize custom tie-breaking priorities per catalog
  • preserve star semantics independently of ranking configuration
  • validate combine modes, priorities, and flag-cut combinations early
  • improve adaptive worker scaling and recovery configuration
  • expand lightweight tests for deduplication and distributed outputs
  • remove obsolete notebooks, fixtures, scripts, and end-to-end tests

- keep Parquet and CSV consolidation distributed until final assembly
- make draw_one deterministic and lazy for large catalogs
- validate and normalize custom tie-breaking priorities per catalog
- preserve star semantics independently of ranking configuration
- validate combine modes, priorities, and flag-cut combinations early
- improve adaptive worker scaling and recovery configuration
- expand lightweight tests for deduplication and distributed outputs
- remove obsolete notebooks, fixtures, scripts, and end-to-end tests
- add checks for YAML, JSON, TOML, merge conflicts, and large files
- detect private keys, debug statements, and case conflicts
- prevent direct commits to the main branch
- enable automatic Ruff import sorting
- normalize existing imports, whitespace, and file endings
@luigilcsilva
luigilcsilva merged commit c1dd71f into main Jul 6, 2026
2 checks passed
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.

Strengthen distributed deduplication, validation, and observability in the CRC pipeline

1 participant