thwap/kronai

Watch

Fork 0

feat(discovery): implement job discovery system #77

Merged

fuzzy merged 3 commits from feature/0014-implement-job-discovery into main

2026-07-05 13:48:31 +00:00

automation commented

2026-07-05 13:40:23 +00:00

Owner

What

Implement job discovery: scan interval directories, detect job type (ephemeral vs recurring), parse numeric IDs, and load job configs with defaults.

Why

Enables the discovery layer that feeds into the job execution engine.

Testing

24 unit tests pass
Pre-commit hooks pass (ruff, mypy, formatting)
Coverage: config 100%, discovery 100%

Breaking Changes

None.

Notes

Job type determined by marker files: .ephemeral = run-once/delete, .recurring (or absent) = keep after run. Non-numeric directories are skipped with a warning.

Closes #0010
Closes #0011
Closes #0012
Closes #0013
Closes #0014

## What Implement job discovery: scan interval directories, detect job type (ephemeral vs recurring), parse numeric IDs, and load job configs with defaults. ## Why Enables the discovery layer that feeds into the job execution engine. ## Testing - [x] 24 unit tests pass - [x] Pre-commit hooks pass (ruff, mypy, formatting) - [x] Coverage: config 100%, discovery 100% ## Breaking Changes None. ## Notes Job type determined by marker files: `.ephemeral` = run-once/delete, `.recurring` (or absent) = keep after run. Non-numeric directories are skipped with a warning. Closes #0010 Closes #0011 Closes #0012 Closes #0013 Closes #0014

automation added 3 commits

2026-07-05 13:40:23 +00:00

feat(discovery): implement job discovery system 6d44d10292

- discover_jobs: orchestrate scanning, type detection, config loading
- scan_intervals: find hourly/daily/weekly/monthly dirs with job IDs
- detect_job_type: marker files (.ephemeral / .recurring) for lifecycle
- parse_job_id: extract numeric ID from directory name
- load_job_config: read job config.toml with default fallback
Closes #0010
Closes #0011
Closes #0012
Closes #0013

test(discovery): add unit tests for job discovery 010d313463

Cover interval scanning, ID parsing, type detection, config loading,
full discovery, and error cases.

docs: mark Phase 1 Task 3 complete

CI / lint (pull_request) Successful in 1m26s

Details

CI / test (pull_request) Successful in 49s

Details

9924805d86

Job discovery with interval scanning, type detection, ID parsing,
and default config handling done.
Closes #0014

the.auditor approved these changes

2026-07-05 13:46:48 +00:00

the.auditor left a comment

Automated Review — PR #77

Result: APPROVED

Clean implementation of the discovery layer. Here's the analysis:

Correctness: Directory structure scanning, ID parsing, type detection, and config loading all work as described. Non-numeric directories are skipped with a warning. Missing config falls back to defaults. Marker file detection is clear.
Completeness: Closes #10–#14. ROADMAP updated.
Test coverage: 24 tests, well-organized into test classes, 100% discovery module coverage. Tests cover empty intervals, unknown folders, non-numeric dirs, marker files, partial config, and error paths.
Error handling: DiscoveryError for missing jobs directory. ValueError from parse_job_id caught per-job with warning. ConfigError from invalid config.toml propagates up (see suggestion #3).
Maintainability: Clean code, constants at module level, good separation of concerns between scan/parse/detect/load.

Suggestions (non-blocking)

src/kronai/discovery.py:66-70 — except ValueError wraps three function calls but only parse_job_id raises it. Consider narrowing the try/except to just the parse_job_id() call to avoid silently swallowing future ValueError from other functions:

try:
    job_id = parse_job_id(job_path)
except ValueError:
    logger.warning("Skipping non-numeric directory: %s", job_path)
    continue
job_type = detect_job_type(job_path)
config = load_job_config(job_path)

src/kronai/discovery.py:130-137 — raw.get() returns object but values flow into typed JobConfig() fields. e.g., mcp_servers expects list[str] but a TOML table or scalar would slip through silently. Consider adding isinstance guards or using a Pydantic-style validator in the future.
src/kronai/discovery.py — Invalid config.toml raises ConfigError which propagates from discover_jobs() and halts discovery of all jobs. Consider catching it per-job (log warning, use defaults) so one bad config doesn't block everything.

## Automated Review — PR #77 **Result: APPROVED** --- Clean implementation of the discovery layer. Here's the analysis: - **Correctness**: Directory structure scanning, ID parsing, type detection, and config loading all work as described. Non-numeric directories are skipped with a warning. Missing config falls back to defaults. Marker file detection is clear. - **Completeness**: Closes #10–#14. ROADMAP updated. - **Test coverage**: 24 tests, well-organized into test classes, 100% discovery module coverage. Tests cover empty intervals, unknown folders, non-numeric dirs, marker files, partial config, and error paths. - **Error handling**: `DiscoveryError` for missing jobs directory. `ValueError` from `parse_job_id` caught per-job with warning. `ConfigError` from invalid config.toml propagates up (see suggestion #3). - **Maintainability**: Clean code, constants at module level, good separation of concerns between scan/parse/detect/load. ### Suggestions (non-blocking) 1. **`src/kronai/discovery.py:66-70`** — `except ValueError` wraps three function calls but only `parse_job_id` raises it. Consider narrowing the try/except to just the `parse_job_id()` call to avoid silently swallowing future `ValueError` from other functions: ```python try: job_id = parse_job_id(job_path) except ValueError: logger.warning("Skipping non-numeric directory: %s", job_path) continue job_type = detect_job_type(job_path) config = load_job_config(job_path) ``` 2. **`src/kronai/discovery.py:130-137`** — `raw.get()` returns `object` but values flow into typed `JobConfig()` fields. e.g., `mcp_servers` expects `list[str]` but a TOML table or scalar would slip through silently. Consider adding `isinstance` guards or using a Pydantic-style validator in the future. 3. **`src/kronai/discovery.py`** — Invalid `config.toml` raises `ConfigError` which propagates from `discover_jobs()` and halts discovery of all jobs. Consider catching it per-job (log warning, use defaults) so one bad config doesn't block everything.

the.auditor referenced this pull request

2026-07-05 13:48:01 +00:00

discovery: narrow ValueError catch to only parse_job_id #78

the.auditor referenced this pull request

2026-07-05 13:48:02 +00:00

discovery: load_job_config type safety for raw.get() values #79

the.auditor referenced this pull request

2026-07-05 13:48:03 +00:00

discovery: handle invalid job config.toml gracefully #80

fuzzy merged commit 9924805d86 into main

2026-07-05 13:48:31 +00:00

fuzzy deleted branch feature/0014-implement-job-discovery