Skip to content

Cache or pre-seed FalkorDB test data #551

Description

@gkorland

1. Cache or pre-seed FalkorDB test data (saves ~3.5 min/shard = ~7 min total)

seed_test_data.py clones 2 GitHub repos (GraphRAG-SDK, Flask) and runs full source analysis every run. This is the single biggest time sink at 37% of Playwright runtime.

Options:

  • Best: Export the seeded graph as an RDB dump, commit it as a test fixture, and restore with redis-cli. Eliminates the 220s step entirely.
  • Good: Cache the cloned repos + analysis output with actions/cache keyed on the seed script hash + repo commit SHAs.
  • Minimum: Cache just the git clones to skip network time.

Metadata

Metadata

Labels

No labels
No labels

Type

No type
No fields configured for issues without a type.

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions