Skip to content

fix: create .graph_hashes.json after init/reprocess#298

Merged
HumanBean17 merged 1 commit into
masterfrom
feat/init-creates-graph-hashes
Jun 8, 2026
Merged

fix: create .graph_hashes.json after init/reprocess#298
HumanBean17 merged 1 commit into
masterfrom
feat/init-creates-graph-hashes

Conversation

@HumanBean17

Copy link
Copy Markdown
Owner

Summary

  • Move _init_hash_tracker() call into write_kuzu() so init and reprocess persist .graph_hashes.json after building the graph
  • Remove two now-redundant _init_hash_tracker() calls from incremental fallback paths (incremental_rebuild and _fallback_to_full)

Problem

After init or reprocess, the first increment command treated all files as new because .graph_hashes.json didn't exist. This caused an unnecessary full rebuild every time.

Verification

rm -rf /tmp/check && .venv/bin/python build_ast_graph.py \
  --source-root tests/bank-chat-system \
  --kuzu-path /tmp/check/code_graph.kuzu --verbose
ls /tmp/check/.graph_hashes.json  # now exists with 130 files hashed
.venv/bin/ruff check .
.venv/bin/python -m pytest tests -v  # 741 passed, 11 skipped

Closes #297

🤖 Generated with Claude Code

…t is fast

Move _init_hash_tracker() call into write_kuzu() so that init and reprocess
commands persist file hashes alongside the graph. Previously the first
increment after init/reprocess treated all files as new, causing an
unnecessary full rebuild.

Closes #297

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
@HumanBean17 HumanBean17 merged commit b461c90 into master Jun 8, 2026
1 check failed
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.

Design gap: and don't create .graph_hashes.json, causing first to be slow

1 participant