Repository navigation
v0.3.0 Added SQLite Database Backend
1. Key Improvements
-
Database Backend (
procdb.py)- Added SQLite-based process tracking (replaces JSON files)
- Structured schema with
uuid,pid,name,cmd(JSON), and timestamps - Better process lifecycle management
-
Process Management (
spawn.py)- New
Spawnclass with enhanced process isolation - Added
splitoption to separate stdout/stderr logging - Improved PID handling and cleanup
- New
-
CLI Enhancements (
cli.py)- New
--splitflag for separate stdout/stderr capture - Improved status formatting (
✅ Live/❌ Gone) - Better error handling for ambiguous/nonexistent processes
- New
-
Utility Functions (
utils.py)- Added
generate_pseudowords()for default process names - Enhanced
look_multiple()for partial/fuzzy matching - Case-insensitive slug generation
- Added
2. Breaking Changes
- Storage Format: Migrated from JSON files to SQLite database (
/tmp/runce.v1/db) - Status Labels: Changed from
✅ Running/👻 Absentto✅ Live/❌ Gone
3. New Features
-
Split Output Logging:
runce run --split --id my-task -- python script.py
-
Better Process Matching:
- Exact names (
runce kill my-task) - Partial matches (
runce kill my-) - Ambiguous detection (multiple matches)
- Exact names (
-
Improved CLI Output:
Status Elapsed PID Name, Command ─────── ──────── ─────── ──────────── ✅ Live 00:01:23 1234 my-task, python script.py
4. Bug Fixes
- Fixed PID cleanup race conditions
- Improved error handling for duplicate processes
- Better handling of subprocess stdin/stdout/stderr
5. Test Coverage
- Added
test_cli.pyfor end-to-end CLI testing - Expanded
test_utils.pywith new utility checks - New
test_look.pyfor process matching logic
6. Example Workflow
# Start process with split logging
runce run --split --id my-script -- python script.py
# Check status (new format)
runce list
# Tail stderr only
runce tail --err my-script
# Clean up
runce kill --remove my-script