When validateJSONLIntegrity() clears export_hashes due to hash mismatch
or missing JSONL, the subsequent export now correctly exports ALL issues
instead of only dirty ones, preventing permanent database divergence.
Changes:
- validateJSONLIntegrity() returns (needsFullExport, error) to signal when
export_hashes was cleared
- flushToJSONL() moved integrity check BEFORE isDirty gate so integrity
issues trigger export even when nothing is dirty
- Missing JSONL treated as non-fatal force-full-export case
- Increased scanner buffer from 64KB to 2MB to handle large JSON lines
- Added scanner.Err() check to catch buffer overflow errors
- Updated all tests to verify needsFullExport flag
Fixes database divergence issue where clearing export_hashes didn't
trigger re-export, causing 5 issues to disappear from JSONL in fred clone.
Amp-Thread-ID: https://ampcode.com/threads/T-bf2fdcd6-7bbd-4c30-b1db-746b928c93b8
Co-authored-by: Amp <amp@ampcode.com>
- Added checkDependencyCycles() function using recursive CTE
- Integrated as Check #10 in runDiagnostics()
- Reports error status with count and fix suggestion if cycles found
- Updated doctor command documentation
- Fixes bd-3e3b
When using `bd list --json`, each issue now includes:
- `dependency_count`: Number of issues this issue depends on
- `dependent_count`: Number of issues that depend on this issue
This provides quick access to dependency relationship counts without
needing to fetch full dependency lists or run multiple bd show commands.
Performance:
- Uses single bulk query (GetDependencyCounts) instead of N individual queries
- Overhead: ~26% for 500 issues (24ms vs 19ms baseline)
- Avoids N+1 query problem that would have caused 2.2x slowdown
Implementation:
- Added GetDependencyCounts() to Storage interface for bulk counting
- Efficient SQLite query using UNION ALL + GROUP BY
- Memory storage implementation for testing
- Moved IssueWithCounts to types package to avoid duplication
- Both RPC and direct modes use optimized bulk query
Tests:
- Added comprehensive tests for GetDependencyCounts
- Tests cover: normal operation, empty list, nonexistent IDs
- All existing tests continue to pass
Backwards compatible: JSON structure is additive, all original fields preserved.
Generated with [Claude Code](https://claude.com/claude-code)
Co-authored-by: Claude <noreply@anthropic.com>
- update prefix/number parsing to use the last hyphen across utils and nodb paths
- add regression tests covering multi-hyphen prefixes in both packages
- Use path normalization (EvalSymlinks) to reliably match daemons across symlinks
- Check for stale sockets directly (catches cases where RPC failed)
- Detect multiple daemons for same workspace
- Enhanced error details for stale daemon detection
- Use global daemon registry instead of path-scoped discovery
Test proves the bug exists:
- Agent A closes issue and pushes
- Agent B does git pull (JSONL shows 'closed')
- Daemon query returns 'open' (stale SQLite)
Test currently fails, confirming daemon doesn't auto-import
after git pull updates JSONL. Will pass once fix is complete.
Fixes#197: bd doctor was hardcoded to look for beads.db and didn't
check config.json for custom database names like beady.db.
Now both checkDatabaseVersion() and checkIDFormat() check config.json
first before falling back to the canonical database name.
- Created config.go with Config struct
- Created daemon.go with Daemon struct and Start/Stop methods
- Created logger.go for logging setup
- Created process.go for lock/PID management
- Created fingerprint.go for database validation
- Created flock_unix.go/flock_windows.go for platform-specific locking
- Created git.go for git operations
Still TODO:
- Implement runGlobalDaemon, startRPCServer, runSyncLoop
- Create sync.go, rpc.go, jsonl.go, validation.go
- Update cmd/bd/daemon.go to use daemonrunner
Part of bd-5f26
