kin: KIN-048 Post-pipeline hook: автокоммит после успешного завершения задачи. git add -A && git commit -m 'kin: TASK_ID TITLE'. Срабатывает автоматически как rebuild-frontend.
This commit is contained in:
Gros Frumos
parent
8a6f280cbd
commit
ae21e48b65
13 changed files with 1554 additions and 65 deletions
41
agents/prompts/learner.md
Normal file
41
agents/prompts/learner.md
Normal file
|
|
@ -0,0 +1,41 @@
|
|||
You are a learning extractor for the Kin multi-agent orchestrator.
|
||||
|
||||
Your job: analyze the outputs of a completed pipeline and extract up to 5 valuable pieces of knowledge — architectural decisions, gotchas, or conventions discovered during execution.
|
||||
|
||||
## Input
|
||||
|
||||
You receive:
|
||||
- PIPELINE_OUTPUTS: summary of each step's output (role → first 2000 chars)
|
||||
- EXISTING_DECISIONS: list of already-known decisions (title + type) to avoid duplicates
|
||||
|
||||
## What to extract
|
||||
|
||||
- **decision** — an architectural or design choice made (e.g., "Use UUID for task IDs")
|
||||
- **gotcha** — a pitfall or unexpected problem encountered (e.g., "sqlite3 closes connection on thread switch")
|
||||
- **convention** — a coding or process standard established (e.g., "Always run tests after each change")
|
||||
|
||||
## Rules
|
||||
|
||||
- Extract ONLY genuinely new knowledge not already in EXISTING_DECISIONS
|
||||
- Skip trivial or obvious items (e.g., "write clean code")
|
||||
- Skip task-specific results that won't generalize (e.g., "fixed bug in useSearch.ts line 42")
|
||||
- Each decision must be actionable and reusable across future tasks
|
||||
- Extract at most 5 decisions total; fewer is better than low-quality ones
|
||||
- If nothing valuable found, return empty list
|
||||
|
||||
## Output format
|
||||
|
||||
Return ONLY valid JSON (no markdown, no explanation):
|
||||
|
||||
```json
|
||||
{
|
||||
"decisions": [
|
||||
{
|
||||
"type": "decision",
|
||||
"title": "Short memorable title",
|
||||
"description": "Clear explanation of what was decided and why",
|
||||
"tags": ["optional", "tags"]
|
||||
}
|
||||
]
|
||||
}
|
||||
```
|
||||
|
|
@ -30,6 +30,16 @@ You receive:
|
|||
- Don't assign specialists who aren't needed.
|
||||
- If a task is blocked or unclear, say so — don't guess.
|
||||
|
||||
## Completion mode selection
|
||||
|
||||
Set `completion_mode` based on the following rules (in priority order):
|
||||
|
||||
1. If `project.execution_mode` is set — use it as the default.
|
||||
2. Override by `route_type`:
|
||||
- `debug`, `hotfix`, `feature` → `"auto_complete"` (only if the last pipeline step is `tester` or `reviewer`)
|
||||
- `research`, `new_project`, `security_audit` → `"review"`
|
||||
3. Fallback: `"review"`
|
||||
|
||||
## Output format
|
||||
|
||||
Return ONLY valid JSON (no markdown, no explanation):
|
||||
|
|
@ -37,6 +47,7 @@ Return ONLY valid JSON (no markdown, no explanation):
|
|||
```json
|
||||
{
|
||||
"analysis": "Brief analysis of what needs to be done",
|
||||
"completion_mode": "auto_complete",
|
||||
"pipeline": [
|
||||
{
|
||||
"role": "debugger",
|
||||
|
|
|
|||
252
agents/runner.py
252
agents/runner.py
|
|
@ -4,7 +4,9 @@ Each agent = separate process with isolated context.
|
|||
"""
|
||||
|
||||
import json
|
||||
import logging
|
||||
import os
|
||||
import shutil
|
||||
import sqlite3
|
||||
import subprocess
|
||||
import time
|
||||
|
|
@ -13,6 +15,50 @@ from typing import Any
|
|||
|
||||
import re
|
||||
|
||||
_logger = logging.getLogger("kin.runner")
|
||||
|
||||
|
||||
# Extra PATH entries to inject when searching for claude CLI.
|
||||
# launchctl daemons start with a stripped PATH that may omit these.
|
||||
_EXTRA_PATH_DIRS = [
|
||||
"/opt/homebrew/bin",
|
||||
"/opt/homebrew/sbin",
|
||||
"/usr/local/bin",
|
||||
"/usr/local/sbin",
|
||||
]
|
||||
|
||||
|
||||
def _build_claude_env() -> dict:
|
||||
"""Return an env dict with an extended PATH that includes common CLI tool locations.
|
||||
|
||||
Merges _EXTRA_PATH_DIRS with the current process PATH, deduplicating entries.
|
||||
Also resolves ~/.nvm/versions/node/*/bin globs that launchctl may not expand.
|
||||
"""
|
||||
env = os.environ.copy()
|
||||
existing = env.get("PATH", "").split(":")
|
||||
|
||||
extra = list(_EXTRA_PATH_DIRS)
|
||||
|
||||
# Expand nvm node bin dirs dynamically
|
||||
nvm_root = Path.home() / ".nvm" / "versions" / "node"
|
||||
if nvm_root.is_dir():
|
||||
for node_ver in sorted(nvm_root.iterdir(), reverse=True):
|
||||
bin_dir = node_ver / "bin"
|
||||
if bin_dir.is_dir():
|
||||
extra.append(str(bin_dir))
|
||||
|
||||
seen = set(existing)
|
||||
new_dirs = [d for d in extra if d and d not in seen]
|
||||
env["PATH"] = ":".join(new_dirs + existing)
|
||||
return env
|
||||
|
||||
|
||||
def _resolve_claude_cmd() -> str:
|
||||
"""Return the full path to the claude CLI, or 'claude' as fallback."""
|
||||
extended_env = _build_claude_env()
|
||||
found = shutil.which("claude", path=extended_env["PATH"])
|
||||
return found or "claude"
|
||||
|
||||
from core import models
|
||||
from core.context_builder import build_context, format_prompt
|
||||
from core.hooks import run_hooks
|
||||
|
|
@ -116,10 +162,12 @@ def _run_claude(
|
|||
working_dir: str | None = None,
|
||||
allow_write: bool = False,
|
||||
noninteractive: bool = False,
|
||||
timeout: int | None = None,
|
||||
) -> dict:
|
||||
"""Execute claude CLI as subprocess. Returns dict with output, returncode, etc."""
|
||||
claude_cmd = _resolve_claude_cmd()
|
||||
cmd = [
|
||||
"claude",
|
||||
claude_cmd,
|
||||
"-p", prompt,
|
||||
"--output-format", "json",
|
||||
"--model", model,
|
||||
|
|
@ -128,7 +176,9 @@ def _run_claude(
|
|||
cmd.append("--dangerously-skip-permissions")
|
||||
|
||||
is_noninteractive = noninteractive or os.environ.get("KIN_NONINTERACTIVE") == "1"
|
||||
timeout = 300 if is_noninteractive else 600
|
||||
if timeout is None:
|
||||
timeout = int(os.environ.get("KIN_AGENT_TIMEOUT") or 600)
|
||||
env = _build_claude_env()
|
||||
|
||||
try:
|
||||
proc = subprocess.run(
|
||||
|
|
@ -137,6 +187,7 @@ def _run_claude(
|
|||
text=True,
|
||||
timeout=timeout,
|
||||
cwd=working_dir,
|
||||
env=env,
|
||||
stdin=subprocess.DEVNULL if is_noninteractive else None,
|
||||
)
|
||||
except FileNotFoundError:
|
||||
|
|
@ -377,6 +428,179 @@ def _is_permission_error(result: dict) -> bool:
|
|||
return any(re.search(p, text) for p in PERMISSION_PATTERNS)
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Autocommit: git add -A && git commit after successful pipeline
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
def _run_autocommit(
|
||||
conn: sqlite3.Connection,
|
||||
task_id: str,
|
||||
project_id: str,
|
||||
) -> None:
|
||||
"""Auto-commit changes after successful pipeline completion.
|
||||
|
||||
Runs: git add -A && git commit -m 'kin: {task_id} {title}'.
|
||||
Silently skips if nothing to commit (exit code 1) or project path not found.
|
||||
Never raises — autocommit errors must never block the pipeline.
|
||||
Uses stderr=subprocess.DEVNULL per decision #30.
|
||||
"""
|
||||
task = models.get_task(conn, task_id)
|
||||
project = models.get_project(conn, project_id)
|
||||
if not task or not project:
|
||||
return
|
||||
|
||||
if not project.get("autocommit_enabled"):
|
||||
return
|
||||
|
||||
project_path = Path(project["path"]).expanduser()
|
||||
if not project_path.is_dir():
|
||||
return
|
||||
|
||||
working_dir = str(project_path)
|
||||
env = _build_claude_env()
|
||||
git_cmd = shutil.which("git", path=env["PATH"]) or "git"
|
||||
|
||||
title = (task.get("title") or "").replace('"', "'").replace("\n", " ").replace("\r", "")
|
||||
commit_msg = f"kin: {task_id} {title}"
|
||||
|
||||
try:
|
||||
subprocess.run(
|
||||
[git_cmd, "add", "-A"],
|
||||
cwd=working_dir,
|
||||
env=env,
|
||||
stderr=subprocess.DEVNULL,
|
||||
stdout=subprocess.DEVNULL,
|
||||
)
|
||||
result = subprocess.run(
|
||||
[git_cmd, "commit", "-m", commit_msg],
|
||||
cwd=working_dir,
|
||||
env=env,
|
||||
stderr=subprocess.DEVNULL,
|
||||
stdout=subprocess.DEVNULL,
|
||||
)
|
||||
if result.returncode == 0:
|
||||
_logger.info("Autocommit: %s", commit_msg)
|
||||
else:
|
||||
_logger.debug("Autocommit: nothing to commit for %s", task_id)
|
||||
except Exception as exc:
|
||||
_logger.warning("Autocommit failed for %s: %s", task_id, exc)
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Auto-learning: extract decisions from pipeline results
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
VALID_DECISION_TYPES = {"decision", "gotcha", "convention"}
|
||||
|
||||
def _run_learning_extraction(
|
||||
conn: sqlite3.Connection,
|
||||
task_id: str,
|
||||
project_id: str,
|
||||
step_results: list[dict],
|
||||
) -> dict:
|
||||
"""Extract and save decisions from completed pipeline results.
|
||||
|
||||
Calls the learner agent with step outputs + existing decisions,
|
||||
parses the JSON response, and saves new decisions via add_decision_if_new.
|
||||
Returns a summary dict with added/skipped counts.
|
||||
"""
|
||||
learner_prompt_path = PROMPTS_DIR / "learner.md"
|
||||
if not learner_prompt_path.exists():
|
||||
return {"added": 0, "skipped": 0, "error": "learner.md not found"}
|
||||
|
||||
template = learner_prompt_path.read_text()
|
||||
|
||||
# Summarize step outputs (first 2000 chars each)
|
||||
step_summaries = {}
|
||||
for r in step_results:
|
||||
role = r.get("role", "unknown")
|
||||
output = r.get("raw_output") or r.get("output") or ""
|
||||
if isinstance(output, (dict, list)):
|
||||
output = json.dumps(output, ensure_ascii=False)
|
||||
step_summaries[role] = output[:2000]
|
||||
|
||||
# Fetch existing decisions for dedup hint
|
||||
existing = models.get_decisions(conn, project_id)
|
||||
existing_hints = [
|
||||
{"title": d["title"], "type": d["type"]}
|
||||
for d in existing
|
||||
]
|
||||
|
||||
prompt_parts = [
|
||||
template,
|
||||
"",
|
||||
"## PIPELINE_OUTPUTS",
|
||||
json.dumps(step_summaries, ensure_ascii=False, indent=2),
|
||||
"",
|
||||
"## EXISTING_DECISIONS",
|
||||
json.dumps(existing_hints, ensure_ascii=False, indent=2),
|
||||
]
|
||||
prompt = "\n".join(prompt_parts)
|
||||
|
||||
learner_timeout = int(os.environ.get("KIN_LEARNER_TIMEOUT") or 120)
|
||||
start = time.monotonic()
|
||||
result = _run_claude(prompt, model="sonnet", noninteractive=True, timeout=learner_timeout)
|
||||
duration = int(time.monotonic() - start)
|
||||
|
||||
raw_output = result.get("output", "")
|
||||
if not isinstance(raw_output, str):
|
||||
raw_output = json.dumps(raw_output, ensure_ascii=False)
|
||||
success = result["returncode"] == 0
|
||||
|
||||
# Log to agent_logs
|
||||
models.log_agent_run(
|
||||
conn,
|
||||
project_id=project_id,
|
||||
task_id=task_id,
|
||||
agent_role="learner",
|
||||
action="learn",
|
||||
input_summary=f"project={project_id}, task={task_id}, steps={len(step_results)}",
|
||||
output_summary=raw_output or None,
|
||||
tokens_used=result.get("tokens_used"),
|
||||
model="sonnet",
|
||||
cost_usd=result.get("cost_usd"),
|
||||
success=success,
|
||||
error_message=result.get("error") if not success else None,
|
||||
duration_seconds=duration,
|
||||
)
|
||||
|
||||
parsed = _try_parse_json(raw_output)
|
||||
if not isinstance(parsed, dict):
|
||||
return {"added": 0, "skipped": 0, "error": "non-JSON learner output"}
|
||||
|
||||
decisions = parsed.get("decisions", [])
|
||||
if not isinstance(decisions, list):
|
||||
return {"added": 0, "skipped": 0, "error": "invalid decisions format"}
|
||||
|
||||
added = 0
|
||||
skipped = 0
|
||||
for item in decisions[:5]:
|
||||
if not isinstance(item, dict):
|
||||
continue
|
||||
d_type = item.get("type", "decision")
|
||||
if d_type not in VALID_DECISION_TYPES:
|
||||
d_type = "decision"
|
||||
d_title = (item.get("title") or "").strip()
|
||||
d_desc = (item.get("description") or "").strip()
|
||||
if not d_title or not d_desc:
|
||||
continue
|
||||
saved = models.add_decision_if_new(
|
||||
conn,
|
||||
project_id=project_id,
|
||||
type=d_type,
|
||||
title=d_title,
|
||||
description=d_desc,
|
||||
tags=item.get("tags") or [],
|
||||
task_id=task_id,
|
||||
)
|
||||
if saved:
|
||||
added += 1
|
||||
else:
|
||||
skipped += 1
|
||||
|
||||
return {"added": added, "skipped": skipped}
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Pipeline executor
|
||||
# ---------------------------------------------------------------------------
|
||||
|
|
@ -485,7 +709,7 @@ def run_pipeline(
|
|||
|
||||
if not result["success"]:
|
||||
# Auto mode: retry once with allow_write on permission error
|
||||
if mode == "auto" and not allow_write and _is_permission_error(result):
|
||||
if mode == "auto_complete" and not allow_write and _is_permission_error(result):
|
||||
task_modules = models.get_modules(conn, project_id)
|
||||
try:
|
||||
run_hooks(conn, project_id, task_id,
|
||||
|
|
@ -555,8 +779,11 @@ def run_pipeline(
|
|||
|
||||
task_modules = models.get_modules(conn, project_id)
|
||||
|
||||
if mode == "auto":
|
||||
# Auto mode: skip review, approve immediately
|
||||
last_role = steps[-1].get("role", "") if steps else ""
|
||||
auto_eligible = last_role in {"tester", "reviewer"}
|
||||
|
||||
if mode == "auto_complete" and auto_eligible:
|
||||
# Auto-complete mode: last step is tester/reviewer — skip review, approve immediately
|
||||
models.update_task(conn, task_id, status="done")
|
||||
try:
|
||||
run_hooks(conn, project_id, task_id,
|
||||
|
|
@ -586,7 +813,7 @@ def run_pipeline(
|
|||
pass
|
||||
else:
|
||||
# Review mode: wait for manual approval
|
||||
models.update_task(conn, task_id, status="review")
|
||||
models.update_task(conn, task_id, status="review", execution_mode="review")
|
||||
|
||||
# Run post-pipeline hooks (failures don't affect pipeline status)
|
||||
try:
|
||||
|
|
@ -595,6 +822,19 @@ def run_pipeline(
|
|||
except Exception:
|
||||
pass # Hook errors must never block pipeline completion
|
||||
|
||||
# Auto-learning: extract decisions from pipeline results
|
||||
if results:
|
||||
try:
|
||||
_run_learning_extraction(conn, task_id, project_id, results)
|
||||
except Exception:
|
||||
pass # Learning errors must never block pipeline completion
|
||||
|
||||
# Auto-commit changes after successful pipeline
|
||||
try:
|
||||
_run_autocommit(conn, task_id, project_id)
|
||||
except Exception:
|
||||
pass # Autocommit errors must never block pipeline completion
|
||||
|
||||
return {
|
||||
"success": True,
|
||||
"steps_completed": len(steps),
|
||||
|
|
|
|||
12
cli/main.py
12
cli/main.py
|
|
@ -586,6 +586,18 @@ def run_task(ctx, task_id, dry_run, allow_write):
|
|||
pipeline_steps = output["pipeline"]
|
||||
analysis = output.get("analysis", "")
|
||||
|
||||
# Save completion_mode from PM output to task (only if not already set by user)
|
||||
task_current = models.get_task(conn, task_id)
|
||||
if not task_current.get("execution_mode"):
|
||||
pm_completion_mode = models.validate_completion_mode(
|
||||
output.get("completion_mode", "review")
|
||||
)
|
||||
models.update_task(conn, task_id, execution_mode=pm_completion_mode)
|
||||
import logging
|
||||
logging.getLogger("kin").info(
|
||||
"PM set completion_mode=%s for task %s", pm_completion_mode, task_id
|
||||
)
|
||||
|
||||
click.echo(f"\nAnalysis: {analysis}")
|
||||
click.echo(f"Pipeline ({len(pipeline_steps)} steps):")
|
||||
for i, step in enumerate(pipeline_steps, 1):
|
||||
|
|
|
|||
|
|
@ -110,6 +110,7 @@ def _slim_project(project: dict) -> dict:
|
|||
"path": project["path"],
|
||||
"tech_stack": project.get("tech_stack"),
|
||||
"language": project.get("language", "ru"),
|
||||
"execution_mode": project.get("execution_mode"),
|
||||
}
|
||||
|
||||
|
||||
|
|
|
|||
49
core/db.py
49
core/db.py
|
|
@ -216,12 +216,61 @@ def _migrate(conn: sqlite3.Connection):
|
|||
conn.execute("ALTER TABLE tasks ADD COLUMN blocked_reason TEXT")
|
||||
conn.commit()
|
||||
|
||||
if "autocommit_enabled" not in proj_cols:
|
||||
conn.execute("ALTER TABLE projects ADD COLUMN autocommit_enabled INTEGER DEFAULT 0")
|
||||
conn.commit()
|
||||
|
||||
# Rename legacy 'auto' → 'auto_complete' (KIN-063)
|
||||
conn.execute(
|
||||
"UPDATE projects SET execution_mode = 'auto_complete' WHERE execution_mode = 'auto'"
|
||||
)
|
||||
conn.execute(
|
||||
"UPDATE tasks SET execution_mode = 'auto_complete' WHERE execution_mode = 'auto'"
|
||||
)
|
||||
conn.commit()
|
||||
|
||||
|
||||
def _seed_default_hooks(conn: sqlite3.Connection):
|
||||
"""Seed default hooks for the kin project (idempotent).
|
||||
|
||||
Creates rebuild-frontend hook only when:
|
||||
- project 'kin' exists in the projects table
|
||||
- the hook doesn't already exist (no duplicate)
|
||||
"""
|
||||
kin_exists = conn.execute(
|
||||
"SELECT 1 FROM projects WHERE id = 'kin'"
|
||||
).fetchone()
|
||||
if not kin_exists:
|
||||
return
|
||||
|
||||
exists = conn.execute(
|
||||
"SELECT 1 FROM hooks"
|
||||
" WHERE project_id = 'kin'"
|
||||
" AND name = 'rebuild-frontend'"
|
||||
" AND event = 'pipeline_completed'"
|
||||
).fetchone()
|
||||
if not exists:
|
||||
conn.execute(
|
||||
"""INSERT INTO hooks (project_id, name, event, command, enabled)
|
||||
VALUES ('kin', 'rebuild-frontend', 'pipeline_completed',
|
||||
'cd /Users/grosfrumos/projects/kin/web/frontend && npm run build',
|
||||
1)"""
|
||||
)
|
||||
conn.commit()
|
||||
|
||||
# Enable autocommit for kin project (opt-in, idempotent)
|
||||
conn.execute(
|
||||
"UPDATE projects SET autocommit_enabled=1 WHERE id='kin' AND autocommit_enabled=0"
|
||||
)
|
||||
conn.commit()
|
||||
|
||||
|
||||
def init_db(db_path: Path = DB_PATH) -> sqlite3.Connection:
|
||||
conn = get_connection(db_path)
|
||||
conn.executescript(SCHEMA)
|
||||
conn.commit()
|
||||
_migrate(conn)
|
||||
_seed_default_hooks(conn)
|
||||
return conn
|
||||
|
||||
|
||||
|
|
|
|||
|
|
@ -14,6 +14,15 @@ VALID_TASK_STATUSES = [
|
|||
"blocked", "decomposed", "cancelled",
|
||||
]
|
||||
|
||||
VALID_COMPLETION_MODES = {"auto_complete", "review"}
|
||||
|
||||
|
||||
def validate_completion_mode(value: str) -> str:
|
||||
"""Validate completion mode from LLM output. Falls back to 'review' if invalid."""
|
||||
if value in VALID_COMPLETION_MODES:
|
||||
return value
|
||||
return "review"
|
||||
|
||||
|
||||
def _row_to_dict(row: sqlite3.Row | None) -> dict | None:
|
||||
"""Convert sqlite3.Row to dict with JSON fields decoded."""
|
||||
|
|
@ -220,6 +229,32 @@ def add_decision(
|
|||
return _row_to_dict(row)
|
||||
|
||||
|
||||
def add_decision_if_new(
|
||||
conn: sqlite3.Connection,
|
||||
project_id: str,
|
||||
type: str,
|
||||
title: str,
|
||||
description: str,
|
||||
category: str | None = None,
|
||||
tags: list | None = None,
|
||||
task_id: str | None = None,
|
||||
) -> dict | None:
|
||||
"""Add a decision only if no existing one matches (project_id, type, normalized title).
|
||||
|
||||
Returns the new decision dict, or None if skipped as duplicate.
|
||||
"""
|
||||
existing = conn.execute(
|
||||
"""SELECT id FROM decisions
|
||||
WHERE project_id = ? AND type = ?
|
||||
AND lower(trim(title)) = lower(trim(?))""",
|
||||
(project_id, type, title),
|
||||
).fetchone()
|
||||
if existing:
|
||||
return None
|
||||
return add_decision(conn, project_id, type, title, description,
|
||||
category=category, tags=tags, task_id=task_id)
|
||||
|
||||
|
||||
def get_decisions(
|
||||
conn: sqlite3.Connection,
|
||||
project_id: str,
|
||||
|
|
|
|||
|
|
@ -342,6 +342,19 @@ def test_patch_task_empty_body_returns_400(client):
|
|||
assert r.status_code == 400
|
||||
|
||||
|
||||
def test_patch_task_execution_mode_auto_complete_accepted(client):
|
||||
"""KIN-063: execution_mode='auto_complete' принимается (200)."""
|
||||
r = client.patch("/api/tasks/P1-001", json={"execution_mode": "auto_complete"})
|
||||
assert r.status_code == 200
|
||||
assert r.json()["execution_mode"] == "auto_complete"
|
||||
|
||||
|
||||
def test_patch_task_execution_mode_auto_rejected(client):
|
||||
"""KIN-063: старое значение 'auto' должно отклоняться (400) — Decision #29."""
|
||||
r = client.patch("/api/tasks/P1-001", json={"execution_mode": "auto"})
|
||||
assert r.status_code == 400
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# KIN-022 — blocked_reason: регрессионные тесты
|
||||
# ---------------------------------------------------------------------------
|
||||
|
|
@ -589,3 +602,27 @@ def test_run_kin_040_allow_write_true_ignored(client):
|
|||
Эндпоинт не имеет body-параметра, поэтому FastAPI не валидирует тело."""
|
||||
r = client.post("/api/tasks/P1-001/run", json={"allow_write": True})
|
||||
assert r.status_code == 202
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# KIN-058 — регрессионный тест: stderr=DEVNULL у Popen в web API
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
def test_run_sets_stderr_devnull(client):
|
||||
"""Регрессионный тест KIN-058: stderr=DEVNULL всегда устанавливается в Popen,
|
||||
чтобы stderr дочернего процесса не загрязнял логи uvicorn."""
|
||||
import subprocess as _subprocess
|
||||
from unittest.mock import patch, MagicMock
|
||||
with patch("web.api.subprocess.Popen") as mock_popen:
|
||||
mock_proc = MagicMock()
|
||||
mock_proc.pid = 77
|
||||
mock_popen.return_value = mock_proc
|
||||
|
||||
r = client.post("/api/tasks/P1-001/run")
|
||||
assert r.status_code == 202
|
||||
|
||||
call_kwargs = mock_popen.call_args[1]
|
||||
assert call_kwargs.get("stderr") == _subprocess.DEVNULL, (
|
||||
"Регрессия KIN-058: stderr у Popen должен быть DEVNULL, "
|
||||
"иначе вывод агента попадает в логи uvicorn"
|
||||
)
|
||||
|
|
|
|||
|
|
@ -1,7 +1,8 @@
|
|||
"""
|
||||
Tests for KIN-012 auto mode features:
|
||||
Tests for KIN-012/KIN-063 auto mode features:
|
||||
|
||||
- TestAutoApprove: pipeline auto-approves (status → done) без ручного review
|
||||
(KIN-063: auto_complete только если последний шаг — tester или reviewer)
|
||||
- TestAutoRerunOnPermissionDenied: runner делает retry при permission error,
|
||||
останавливается после одного retry (лимит = 1)
|
||||
- TestAutoFollowup: generate_followups вызывается сразу, без ожидания
|
||||
|
|
@ -75,30 +76,30 @@ class TestAutoApprove:
|
|||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_auto_mode_sets_status_done(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""Auto-режим: статус задачи становится 'done', а не 'review'."""
|
||||
"""Auto-complete режим: статус становится 'done', если последний шаг — tester."""
|
||||
mock_run.return_value = _mock_success()
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
steps = [{"role": "debugger", "brief": "find bug"}]
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "find bug"}, {"role": "tester", "brief": "verify fix"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
task = models.get_task(conn, "VDOL-001")
|
||||
assert task["status"] == "done", "Auto-mode должен auto-approve: status=done"
|
||||
assert task["status"] == "done", "Auto-complete должен auto-approve: status=done"
|
||||
|
||||
@patch("core.followup.generate_followups")
|
||||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_auto_mode_fires_task_auto_approved_hook(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""В auto-режиме срабатывает хук task_auto_approved."""
|
||||
"""В auto_complete-режиме срабатывает хук task_auto_approved (если последний шаг — tester)."""
|
||||
mock_run.return_value = _mock_success()
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
steps = [{"role": "debugger", "brief": "find bug"}]
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "find bug"}, {"role": "tester", "brief": "verify"}]
|
||||
run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
events = _get_hook_events(mock_hooks)
|
||||
|
|
@ -140,20 +141,20 @@ class TestAutoApprove:
|
|||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_task_level_auto_overrides_project_review(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""Если у задачи execution_mode=auto, pipeline auto-approve, даже если проект в review."""
|
||||
"""Если у задачи execution_mode=auto_complete, pipeline auto-approve, даже если проект в review."""
|
||||
mock_run.return_value = _mock_success()
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
# Проект в review, но задача — auto
|
||||
models.update_task(conn, "VDOL-001", execution_mode="auto")
|
||||
# Проект в review, но задача — auto_complete
|
||||
models.update_task(conn, "VDOL-001", execution_mode="auto_complete")
|
||||
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
steps = [{"role": "debugger", "brief": "find"}, {"role": "reviewer", "brief": "approve"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
task = models.get_task(conn, "VDOL-001")
|
||||
assert task["status"] == "done", "Task-level auto должен override project review"
|
||||
assert task["status"] == "done", "Task-level auto_complete должен override project review"
|
||||
|
||||
@patch("core.followup.generate_followups")
|
||||
@patch("agents.runner.run_hooks")
|
||||
|
|
@ -164,11 +165,11 @@ class TestAutoApprove:
|
|||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "find"}, {"role": "tester", "brief": "test"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result.get("mode") == "auto"
|
||||
assert result.get("mode") == "auto_complete"
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
|
|
@ -178,10 +179,12 @@ class TestAutoApprove:
|
|||
class TestAutoRerunOnPermissionDenied:
|
||||
"""Runner повторяет шаг при permission issues, останавливается по лимиту (1 retry)."""
|
||||
|
||||
@patch("agents.runner._run_autocommit")
|
||||
@patch("agents.runner._run_learning_extraction")
|
||||
@patch("core.followup.generate_followups")
|
||||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_auto_mode_retries_on_permission_error(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
def test_auto_mode_retries_on_permission_error(self, mock_run, mock_hooks, mock_followup, mock_learn, mock_autocommit, conn):
|
||||
"""Auto-режим: при permission denied runner делает 1 retry с allow_write=True."""
|
||||
mock_run.side_effect = [
|
||||
_mock_permission_denied(), # 1-й вызов: permission error
|
||||
|
|
@ -189,8 +192,9 @@ class TestAutoRerunOnPermissionDenied:
|
|||
]
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
mock_learn.return_value = {"added": 0, "skipped": 0}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "fix file"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
|
|
@ -209,7 +213,7 @@ class TestAutoRerunOnPermissionDenied:
|
|||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "fix"}]
|
||||
run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
|
|
@ -229,7 +233,7 @@ class TestAutoRerunOnPermissionDenied:
|
|||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "fix"}]
|
||||
run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
|
|
@ -248,7 +252,7 @@ class TestAutoRerunOnPermissionDenied:
|
|||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "fix"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
|
|
@ -257,10 +261,12 @@ class TestAutoRerunOnPermissionDenied:
|
|||
task = models.get_task(conn, "VDOL-001")
|
||||
assert task["status"] == "blocked"
|
||||
|
||||
@patch("agents.runner._run_autocommit")
|
||||
@patch("agents.runner._run_learning_extraction")
|
||||
@patch("core.followup.generate_followups")
|
||||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_subsequent_steps_use_allow_write_after_retry(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
def test_subsequent_steps_use_allow_write_after_retry(self, mock_run, mock_hooks, mock_followup, mock_learn, mock_autocommit, conn):
|
||||
"""После успешного retry все следующие шаги тоже используют allow_write."""
|
||||
mock_run.side_effect = [
|
||||
_mock_permission_denied(), # Шаг 1: permission error
|
||||
|
|
@ -269,8 +275,9 @@ class TestAutoRerunOnPermissionDenied:
|
|||
]
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
mock_learn.return_value = {"added": 0, "skipped": 0}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [
|
||||
{"role": "debugger", "brief": "fix"},
|
||||
{"role": "tester", "brief": "test"},
|
||||
|
|
@ -293,7 +300,7 @@ class TestAutoRerunOnPermissionDenied:
|
|||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "fix"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
|
|
@ -330,13 +337,13 @@ class TestAutoFollowup:
|
|||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_auto_followup_triggered_immediately(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""В auto-режиме generate_followups вызывается сразу после pipeline."""
|
||||
"""В auto_complete-режиме generate_followups вызывается сразу после pipeline (последний шаг — tester)."""
|
||||
mock_run.return_value = _mock_success()
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "find"}, {"role": "tester", "brief": "test"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
|
|
@ -357,8 +364,8 @@ class TestAutoFollowup:
|
|||
mock_followup.return_value = {"created": [], "pending_actions": pending}
|
||||
mock_resolve.return_value = [{"resolved": "rerun", "result": {}}]
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "find"}, {"role": "tester", "brief": "test"}]
|
||||
run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
mock_resolve.assert_called_once_with(conn, "VDOL-001", pending)
|
||||
|
|
@ -392,10 +399,10 @@ class TestAutoFollowup:
|
|||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
models.update_task(conn, "VDOL-001", brief={"source": "followup:VDOL-000"})
|
||||
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
steps = [{"role": "debugger", "brief": "find"}, {"role": "tester", "brief": "test"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
|
|
@ -412,8 +419,8 @@ class TestAutoFollowup:
|
|||
mock_hooks.return_value = []
|
||||
mock_followup.side_effect = Exception("followup PM crashed")
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "find"}, {"role": "tester", "brief": "test"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True # Pipeline succeeded, followup failure absorbed
|
||||
|
|
@ -431,8 +438,8 @@ class TestAutoFollowup:
|
|||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
mock_resolve.return_value = []
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "find"}, {"role": "tester", "brief": "test"}]
|
||||
run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
mock_resolve.assert_not_called()
|
||||
|
|
|
|||
|
|
@ -1,6 +1,8 @@
|
|||
"""Tests for core/hooks.py — post-pipeline hook execution."""
|
||||
|
||||
import os
|
||||
import subprocess
|
||||
import tempfile
|
||||
import pytest
|
||||
from unittest.mock import patch, MagicMock
|
||||
|
||||
|
|
@ -539,10 +541,6 @@ class TestKIN052RebuildFrontendCommand:
|
|||
|
||||
Симулирует рестарт: создаём хук, закрываем соединение, открываем новое — хук на месте.
|
||||
"""
|
||||
import tempfile
|
||||
import os
|
||||
from core.db import init_db
|
||||
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
|
|
@ -568,3 +566,109 @@ class TestKIN052RebuildFrontendCommand:
|
|||
assert hooks[0]["trigger_module_path"] is None
|
||||
finally:
|
||||
os.unlink(db_path)
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# KIN-053: _seed_default_hooks — автоматический хук при инициализации БД
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
class TestKIN053SeedDefaultHooks:
|
||||
"""Тесты для _seed_default_hooks (KIN-053).
|
||||
|
||||
При init_db автоматически создаётся rebuild-frontend хук для проекта 'kin',
|
||||
если этот проект уже существует в БД. Функция идемпотентна.
|
||||
"""
|
||||
|
||||
def test_seed_skipped_when_no_kin_project(self):
|
||||
"""_seed_default_hooks не создаёт хук, если проекта 'kin' нет."""
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
conn = init_db(db_path)
|
||||
hooks = get_hooks(conn, "kin", enabled_only=False)
|
||||
conn.close()
|
||||
assert hooks == []
|
||||
finally:
|
||||
os.unlink(db_path)
|
||||
|
||||
def test_seed_creates_hook_when_kin_project_exists(self):
|
||||
"""_seed_default_hooks создаёт rebuild-frontend хук при наличии проекта 'kin'.
|
||||
|
||||
Порядок: init_db → create_project('kin') → повторный init_db → хук есть.
|
||||
"""
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
conn1 = init_db(db_path)
|
||||
models.create_project(conn1, "kin", "Kin", "/projects/kin")
|
||||
conn1.close()
|
||||
|
||||
conn2 = init_db(db_path)
|
||||
hooks = get_hooks(conn2, "kin", event="pipeline_completed", enabled_only=True)
|
||||
conn2.close()
|
||||
|
||||
assert len(hooks) == 1
|
||||
assert hooks[0]["name"] == "rebuild-frontend"
|
||||
assert "npm run build" in hooks[0]["command"]
|
||||
assert "web/frontend" in hooks[0]["command"]
|
||||
finally:
|
||||
os.unlink(db_path)
|
||||
|
||||
def test_seed_hook_has_correct_command(self):
|
||||
"""Команда хука — точная строка с cd && npm run build."""
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
conn1 = init_db(db_path)
|
||||
models.create_project(conn1, "kin", "Kin", "/projects/kin")
|
||||
conn1.close()
|
||||
|
||||
conn2 = init_db(db_path)
|
||||
hooks = get_hooks(conn2, "kin", event="pipeline_completed", enabled_only=False)
|
||||
conn2.close()
|
||||
|
||||
assert hooks[0]["command"] == (
|
||||
"cd /Users/grosfrumos/projects/kin/web/frontend && npm run build"
|
||||
)
|
||||
assert hooks[0]["trigger_module_path"] is None
|
||||
finally:
|
||||
os.unlink(db_path)
|
||||
|
||||
def test_seed_idempotent_no_duplicate(self):
|
||||
"""Повторные вызовы init_db не дублируют хук."""
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
conn = init_db(db_path)
|
||||
models.create_project(conn, "kin", "Kin", "/projects/kin")
|
||||
conn.close()
|
||||
|
||||
for _ in range(3):
|
||||
c = init_db(db_path)
|
||||
c.close()
|
||||
|
||||
conn_final = init_db(db_path)
|
||||
hooks = get_hooks(conn_final, "kin", event="pipeline_completed", enabled_only=False)
|
||||
conn_final.close()
|
||||
|
||||
assert len(hooks) == 1, f"Ожидается 1 хук, получено {len(hooks)}"
|
||||
finally:
|
||||
os.unlink(db_path)
|
||||
|
||||
def test_seed_hook_does_not_affect_other_projects(self):
|
||||
"""Seed не создаёт хуки для других проектов."""
|
||||
with tempfile.NamedTemporaryFile(suffix=".db", delete=False) as f:
|
||||
db_path = f.name
|
||||
try:
|
||||
conn1 = init_db(db_path)
|
||||
models.create_project(conn1, "kin", "Kin", "/projects/kin")
|
||||
models.create_project(conn1, "other", "Other", "/projects/other")
|
||||
conn1.close()
|
||||
|
||||
conn2 = init_db(db_path)
|
||||
other_hooks = get_hooks(conn2, "other", enabled_only=False)
|
||||
conn2.close()
|
||||
|
||||
assert other_hooks == []
|
||||
finally:
|
||||
os.unlink(db_path)
|
||||
|
|
|
|||
|
|
@ -53,6 +53,55 @@ def test_update_project_tech_stack_json(conn):
|
|||
assert updated["tech_stack"] == ["python", "fastapi"]
|
||||
|
||||
|
||||
# -- validate_completion_mode (KIN-063) --
|
||||
|
||||
def test_validate_completion_mode_valid_auto_complete():
|
||||
"""validate_completion_mode принимает 'auto_complete'."""
|
||||
assert models.validate_completion_mode("auto_complete") == "auto_complete"
|
||||
|
||||
|
||||
def test_validate_completion_mode_valid_review():
|
||||
"""validate_completion_mode принимает 'review'."""
|
||||
assert models.validate_completion_mode("review") == "review"
|
||||
|
||||
|
||||
def test_validate_completion_mode_invalid_fallback():
|
||||
"""validate_completion_mode возвращает 'review' для невалидных значений (фоллбэк)."""
|
||||
assert models.validate_completion_mode("auto") == "review"
|
||||
assert models.validate_completion_mode("") == "review"
|
||||
assert models.validate_completion_mode("unknown") == "review"
|
||||
|
||||
|
||||
# -- get_effective_mode (KIN-063) --
|
||||
|
||||
def test_get_effective_mode_task_overrides_project(conn):
|
||||
"""Task execution_mode имеет приоритет над project execution_mode."""
|
||||
models.create_project(conn, "p1", "P1", "/p1", execution_mode="review")
|
||||
models.create_task(conn, "P1-001", "p1", "Task", execution_mode="auto_complete")
|
||||
mode = models.get_effective_mode(conn, "p1", "P1-001")
|
||||
assert mode == "auto_complete"
|
||||
|
||||
|
||||
def test_get_effective_mode_falls_back_to_project(conn):
|
||||
"""Если задача без execution_mode — применяется project execution_mode."""
|
||||
models.create_project(conn, "p1", "P1", "/p1", execution_mode="auto_complete")
|
||||
models.create_task(conn, "P1-001", "p1", "Task") # execution_mode=None
|
||||
mode = models.get_effective_mode(conn, "p1", "P1-001")
|
||||
assert mode == "auto_complete"
|
||||
|
||||
|
||||
def test_get_effective_mode_project_review_overrides_default(conn):
|
||||
"""Project execution_mode='review' + task без override → возвращает 'review'.
|
||||
|
||||
Сценарий: PM хотел auto_complete, но проект настроен на review человеком.
|
||||
get_effective_mode должен вернуть project-level 'review'.
|
||||
"""
|
||||
models.create_project(conn, "p1", "P1", "/p1", execution_mode="review")
|
||||
models.create_task(conn, "P1-001", "p1", "Task") # нет task-level override
|
||||
mode = models.get_effective_mode(conn, "p1", "P1-001")
|
||||
assert mode == "review"
|
||||
|
||||
|
||||
# -- Tasks --
|
||||
|
||||
def test_create_and_get_task(conn):
|
||||
|
|
@ -238,3 +287,46 @@ def test_cost_summary(conn):
|
|||
def test_cost_summary_empty(conn):
|
||||
models.create_project(conn, "p1", "P1", "/p1")
|
||||
assert models.get_cost_summary(conn, days=7) == []
|
||||
|
||||
|
||||
# -- add_decision_if_new --
|
||||
|
||||
def test_add_decision_if_new_adds_new_decision(conn):
|
||||
models.create_project(conn, "p1", "P1", "/p1")
|
||||
d = models.add_decision_if_new(conn, "p1", "gotcha", "Use WAL mode", "description")
|
||||
assert d is not None
|
||||
assert d["title"] == "Use WAL mode"
|
||||
assert d["type"] == "gotcha"
|
||||
|
||||
|
||||
def test_add_decision_if_new_skips_exact_duplicate(conn):
|
||||
models.create_project(conn, "p1", "P1", "/p1")
|
||||
models.add_decision(conn, "p1", "gotcha", "Use WAL mode", "desc1")
|
||||
result = models.add_decision_if_new(conn, "p1", "gotcha", "Use WAL mode", "desc2")
|
||||
assert result is None
|
||||
# Existing decision not duplicated
|
||||
assert len(models.get_decisions(conn, "p1")) == 1
|
||||
|
||||
|
||||
def test_add_decision_if_new_skips_case_insensitive_duplicate(conn):
|
||||
models.create_project(conn, "p1", "P1", "/p1")
|
||||
models.add_decision(conn, "p1", "decision", "Use UUID for task IDs", "desc")
|
||||
result = models.add_decision_if_new(conn, "p1", "decision", "use uuid for task ids", "other desc")
|
||||
assert result is None
|
||||
assert len(models.get_decisions(conn, "p1")) == 1
|
||||
|
||||
|
||||
def test_add_decision_if_new_allows_same_title_different_type(conn):
|
||||
models.create_project(conn, "p1", "P1", "/p1")
|
||||
models.add_decision(conn, "p1", "gotcha", "SQLite WAL", "desc")
|
||||
result = models.add_decision_if_new(conn, "p1", "convention", "SQLite WAL", "other desc")
|
||||
assert result is not None
|
||||
assert len(models.get_decisions(conn, "p1")) == 2
|
||||
|
||||
|
||||
def test_add_decision_if_new_skips_whitespace_duplicate(conn):
|
||||
models.create_project(conn, "p1", "P1", "/p1")
|
||||
models.add_decision(conn, "p1", "convention", "Run tests after each change", "desc")
|
||||
result = models.add_decision_if_new(conn, "p1", "convention", " Run tests after each change ", "desc2")
|
||||
assert result is None
|
||||
assert len(models.get_decisions(conn, "p1")) == 1
|
||||
|
|
|
|||
|
|
@ -6,7 +6,10 @@ import pytest
|
|||
from unittest.mock import patch, MagicMock
|
||||
from core.db import init_db
|
||||
from core import models
|
||||
from agents.runner import run_agent, run_pipeline, run_audit, _try_parse_json
|
||||
from agents.runner import (
|
||||
run_agent, run_pipeline, run_audit, _try_parse_json, _run_learning_extraction,
|
||||
_build_claude_env, _resolve_claude_cmd, _EXTRA_PATH_DIRS, _run_autocommit,
|
||||
)
|
||||
|
||||
|
||||
@pytest.fixture
|
||||
|
|
@ -155,8 +158,9 @@ class TestRunAgent:
|
|||
# ---------------------------------------------------------------------------
|
||||
|
||||
class TestRunPipeline:
|
||||
@patch("agents.runner._run_autocommit") # gotcha #41: мокируем в тестах не о autocommit
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_successful_pipeline(self, mock_run, conn):
|
||||
def test_successful_pipeline(self, mock_run, mock_autocommit, conn):
|
||||
mock_run.return_value = _mock_claude_success({"result": "done"})
|
||||
|
||||
steps = [
|
||||
|
|
@ -298,13 +302,13 @@ class TestAutoMode:
|
|||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_auto_mode_generates_followups(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""Auto mode должен вызывать generate_followups после task_auto_approved."""
|
||||
"""Auto_complete mode должен вызывать generate_followups (последний шаг — tester)."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "done"})
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "find"}, {"role": "tester", "brief": "test"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
|
|
@ -334,15 +338,15 @@ class TestAutoMode:
|
|||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_auto_mode_skips_followups_for_followup_tasks(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""Auto mode НЕ должен генерировать followups для followup-задач (предотвращение рекурсии)."""
|
||||
"""Auto_complete mode НЕ должен генерировать followups для followup-задач (предотвращение рекурсии)."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "done"})
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
models.update_task(conn, "VDOL-001", brief={"source": "followup:VDOL-000"})
|
||||
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
steps = [{"role": "debugger", "brief": "find"}, {"role": "tester", "brief": "test"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
|
|
@ -352,13 +356,13 @@ class TestAutoMode:
|
|||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_auto_mode_fires_task_done_event(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""Auto mode должен вызывать run_hooks с event='task_done' после task_auto_approved."""
|
||||
"""Auto_complete mode должен вызывать run_hooks с event='task_done' (последний шаг — tester)."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "done"})
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "find"}, {"role": "tester", "brief": "test"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
|
|
@ -371,7 +375,7 @@ class TestAutoMode:
|
|||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_auto_mode_resolves_pending_actions(self, mock_run, mock_hooks, mock_followup, mock_resolve, conn):
|
||||
"""Auto mode должен авто-резолвить pending_actions из followup generation."""
|
||||
"""Auto_complete mode должен авто-резолвить pending_actions (последний шаг — tester)."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "done"})
|
||||
mock_hooks.return_value = []
|
||||
|
||||
|
|
@ -380,8 +384,8 @@ class TestAutoMode:
|
|||
mock_followup.return_value = {"created": [], "pending_actions": pending}
|
||||
mock_resolve.return_value = [{"resolved": "rerun", "result": {}}]
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "find"}, {"role": "tester", "brief": "test"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
|
|
@ -393,10 +397,12 @@ class TestAutoMode:
|
|||
# ---------------------------------------------------------------------------
|
||||
|
||||
class TestRetryOnPermissionError:
|
||||
@patch("agents.runner._run_autocommit")
|
||||
@patch("agents.runner._run_learning_extraction")
|
||||
@patch("core.followup.generate_followups")
|
||||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_retry_on_permission_error_auto_mode(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
def test_retry_on_permission_error_auto_mode(self, mock_run, mock_hooks, mock_followup, mock_learn, mock_autocommit, conn):
|
||||
"""Auto mode: retry при permission error должен срабатывать."""
|
||||
permission_fail = _mock_claude_failure("permission denied: cannot write file")
|
||||
retry_success = _mock_claude_success({"result": "fixed"})
|
||||
|
|
@ -404,8 +410,9 @@ class TestRetryOnPermissionError:
|
|||
mock_run.side_effect = [permission_fail, retry_success]
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
mock_learn.return_value = {"added": 0, "skipped": 0}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto")
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
|
|
@ -472,12 +479,13 @@ class TestNonInteractive:
|
|||
call_kwargs = mock_run.call_args[1]
|
||||
assert call_kwargs.get("stdin") == subprocess.DEVNULL
|
||||
|
||||
@patch.dict("os.environ", {"KIN_AGENT_TIMEOUT": ""}, clear=False)
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_noninteractive_uses_300s_timeout(self, mock_run, conn):
|
||||
def test_noninteractive_uses_600s_timeout(self, mock_run, conn):
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
run_agent(conn, "debugger", "VDOL-001", "vdol", noninteractive=True)
|
||||
call_kwargs = mock_run.call_args[1]
|
||||
assert call_kwargs.get("timeout") == 300
|
||||
assert call_kwargs.get("timeout") == 600
|
||||
|
||||
@patch.dict("os.environ", {"KIN_NONINTERACTIVE": ""})
|
||||
@patch("agents.runner.subprocess.run")
|
||||
|
|
@ -504,7 +512,16 @@ class TestNonInteractive:
|
|||
run_agent(conn, "debugger", "VDOL-001", "vdol", noninteractive=False)
|
||||
call_kwargs = mock_run.call_args[1]
|
||||
assert call_kwargs.get("stdin") == subprocess.DEVNULL
|
||||
assert call_kwargs.get("timeout") == 300
|
||||
assert call_kwargs.get("timeout") == 600
|
||||
|
||||
@patch.dict("os.environ", {"KIN_AGENT_TIMEOUT": "900"})
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_custom_timeout_via_env_var(self, mock_run, conn):
|
||||
"""KIN_AGENT_TIMEOUT overrides the default 600s timeout."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
run_agent(conn, "debugger", "VDOL-001", "vdol")
|
||||
call_kwargs = mock_run.call_args[1]
|
||||
assert call_kwargs.get("timeout") == 900
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_allow_write_adds_skip_permissions(self, mock_run, conn):
|
||||
|
|
@ -751,3 +768,786 @@ class TestSilentFailedDiagnostics:
|
|||
|
||||
assert result["success"] is True
|
||||
assert result.get("error") is None
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Auto-learning: _run_learning_extraction
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
class TestRunLearningExtraction:
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_extracts_and_saves_decisions(self, mock_run, conn):
|
||||
"""Успешный сценарий: learner возвращает JSON с decisions, они сохраняются в БД."""
|
||||
learner_output = json.dumps({
|
||||
"decisions": [
|
||||
{"type": "gotcha", "title": "SQLite WAL mode needed", "description": "Without WAL concurrent reads fail", "tags": ["sqlite", "db"]},
|
||||
{"type": "convention", "title": "Always run tests after change", "description": "Prevents regressions", "tags": ["testing"]},
|
||||
]
|
||||
})
|
||||
mock_run.return_value = _mock_claude_success({"result": learner_output})
|
||||
|
||||
step_results = [
|
||||
{"role": "debugger", "raw_output": "Found issue with sqlite concurrent access"},
|
||||
]
|
||||
result = _run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
assert result["added"] == 2
|
||||
assert result["skipped"] == 0
|
||||
|
||||
decisions = conn.execute("SELECT * FROM decisions WHERE project_id='vdol'").fetchall()
|
||||
assert len(decisions) == 2
|
||||
titles = {d["title"] for d in decisions}
|
||||
assert "SQLite WAL mode needed" in titles
|
||||
assert "Always run tests after change" in titles
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_skips_duplicate_decisions(self, mock_run, conn):
|
||||
"""Дедупликация: если decision с таким title+type уже есть, пропускается."""
|
||||
from core import models as m
|
||||
m.add_decision(conn, "vdol", "gotcha", "SQLite WAL mode needed", "existing desc")
|
||||
|
||||
learner_output = json.dumps({
|
||||
"decisions": [
|
||||
{"type": "gotcha", "title": "SQLite WAL mode needed", "description": "duplicate", "tags": []},
|
||||
{"type": "convention", "title": "New convention here", "description": "new desc", "tags": []},
|
||||
]
|
||||
})
|
||||
mock_run.return_value = _mock_claude_success({"result": learner_output})
|
||||
|
||||
step_results = [{"role": "tester", "raw_output": "test output"}]
|
||||
result = _run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
assert result["added"] == 1
|
||||
assert result["skipped"] == 1
|
||||
assert len(conn.execute("SELECT * FROM decisions WHERE project_id='vdol'").fetchall()) == 2
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_limits_to_5_decisions(self, mock_run, conn):
|
||||
"""Learner не должен сохранять более 5 decisions даже если агент вернул больше."""
|
||||
decisions_list = [
|
||||
{"type": "decision", "title": f"Decision {i}", "description": f"desc {i}", "tags": []}
|
||||
for i in range(8)
|
||||
]
|
||||
learner_output = json.dumps({"decisions": decisions_list})
|
||||
mock_run.return_value = _mock_claude_success({"result": learner_output})
|
||||
|
||||
step_results = [{"role": "architect", "raw_output": "long output"}]
|
||||
result = _run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
assert result["added"] == 5
|
||||
assert len(conn.execute("SELECT * FROM decisions WHERE project_id='vdol'").fetchall()) == 5
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_non_json_output_returns_error(self, mock_run, conn):
|
||||
"""Если learner вернул не-JSON, функция возвращает error, не бросает исключение."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "plain text, not json"})
|
||||
|
||||
step_results = [{"role": "debugger", "raw_output": "output"}]
|
||||
result = _run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
assert result["added"] == 0
|
||||
assert "error" in result
|
||||
assert len(conn.execute("SELECT * FROM decisions WHERE project_id='vdol'").fetchall()) == 0
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_decisions_linked_to_task(self, mock_run, conn):
|
||||
"""Сохранённые decisions должны быть привязаны к task_id."""
|
||||
learner_output = json.dumps({
|
||||
"decisions": [
|
||||
{"type": "gotcha", "title": "Important gotcha", "description": "desc", "tags": []},
|
||||
]
|
||||
})
|
||||
mock_run.return_value = _mock_claude_success({"result": learner_output})
|
||||
|
||||
step_results = [{"role": "debugger", "raw_output": "output"}]
|
||||
_run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
d = conn.execute("SELECT * FROM decisions WHERE project_id='vdol'").fetchone()
|
||||
assert d["task_id"] == "VDOL-001"
|
||||
|
||||
@patch("agents.runner._run_learning_extraction")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_pipeline_triggers_learning_after_completion(self, mock_run, mock_learn, conn):
|
||||
"""run_pipeline должен вызывать _run_learning_extraction после успешного завершения."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "done"})
|
||||
mock_learn.return_value = {"added": 1, "skipped": 0}
|
||||
|
||||
steps = [{"role": "debugger", "brief": "find bug"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
mock_learn.assert_called_once()
|
||||
call_args = mock_learn.call_args[0]
|
||||
assert call_args[1] == "VDOL-001" # task_id
|
||||
assert call_args[2] == "vdol" # project_id
|
||||
|
||||
@patch("agents.runner._run_learning_extraction")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_learning_error_does_not_break_pipeline(self, mock_run, mock_learn, conn):
|
||||
"""Если _run_learning_extraction бросает исключение, pipeline не падает."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "done"})
|
||||
mock_learn.side_effect = Exception("learning failed")
|
||||
|
||||
steps = [{"role": "debugger", "brief": "find bug"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
|
||||
def test_pipeline_dry_run_skips_learning(self, conn):
|
||||
"""Dry run не должен вызывать _run_learning_extraction."""
|
||||
steps = [{"role": "debugger", "brief": "find bug"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps, dry_run=True)
|
||||
|
||||
assert result["dry_run"] is True
|
||||
# No decisions saved (dry run — no DB activity)
|
||||
assert len(conn.execute("SELECT * FROM decisions WHERE project_id='vdol'").fetchall()) == 0
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_empty_learner_output_returns_no_decisions(self, mock_run, conn):
|
||||
"""Пустой stdout от learner (subprocess вернул "") — не бросает исключение, возвращает error."""
|
||||
# Используем пустую строку как stdout (не dict), чтобы raw_output оказался пустым
|
||||
mock_run.return_value = _mock_claude_success("")
|
||||
|
||||
step_results = [{"role": "debugger", "raw_output": "output"}]
|
||||
result = _run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
assert result["added"] == 0
|
||||
assert "error" in result
|
||||
assert len(conn.execute("SELECT * FROM decisions WHERE project_id='vdol'").fetchall()) == 0
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_empty_decisions_list_returns_zero_counts(self, mock_run, conn):
|
||||
"""Learner возвращает {"decisions": []} — added=0, skipped=0, без ошибки."""
|
||||
mock_run.return_value = _mock_claude_success({"result": json.dumps({"decisions": []})})
|
||||
|
||||
step_results = [{"role": "debugger", "raw_output": "output"}]
|
||||
result = _run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
assert result["added"] == 0
|
||||
assert result["skipped"] == 0
|
||||
assert "error" not in result
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_decision_missing_title_is_skipped(self, mock_run, conn):
|
||||
"""Decision без title молча пропускается, не вызывает исключение."""
|
||||
learner_output = json.dumps({
|
||||
"decisions": [
|
||||
{"type": "gotcha", "description": "no title here", "tags": []},
|
||||
{"type": "convention", "title": "Valid decision", "description": "desc", "tags": []},
|
||||
]
|
||||
})
|
||||
mock_run.return_value = _mock_claude_success({"result": learner_output})
|
||||
|
||||
step_results = [{"role": "debugger", "raw_output": "output"}]
|
||||
result = _run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
assert result["added"] == 1
|
||||
assert len(conn.execute("SELECT * FROM decisions WHERE project_id='vdol'").fetchall()) == 1
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_decisions_field_not_list_returns_error(self, mock_run, conn):
|
||||
"""Если поле decisions не является списком — возвращается error dict."""
|
||||
mock_run.return_value = _mock_claude_success({"result": json.dumps({"decisions": "not a list"})})
|
||||
|
||||
step_results = [{"role": "debugger", "raw_output": "output"}]
|
||||
result = _run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
assert result["added"] == 0
|
||||
assert "error" in result
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_logs_agent_run_to_db(self, mock_run, conn):
|
||||
"""KIN-060: _run_learning_extraction должна писать запись в agent_logs."""
|
||||
learner_output = json.dumps({
|
||||
"decisions": [
|
||||
{"type": "gotcha", "title": "Log test", "description": "desc", "tags": []},
|
||||
]
|
||||
})
|
||||
mock_run.return_value = _mock_claude_success({"result": learner_output})
|
||||
|
||||
step_results = [{"role": "debugger", "raw_output": "output"}]
|
||||
_run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
logs = conn.execute(
|
||||
"SELECT * FROM agent_logs WHERE agent_role='learner' AND project_id='vdol'"
|
||||
).fetchall()
|
||||
assert len(logs) == 1
|
||||
log = logs[0]
|
||||
assert log["task_id"] == "VDOL-001"
|
||||
assert log["action"] == "learn"
|
||||
assert log["model"] == "sonnet"
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_learner_cost_included_in_cost_summary(self, mock_run, conn):
|
||||
"""KIN-060: get_cost_summary() включает затраты learner-агента."""
|
||||
learner_output = json.dumps({"decisions": []})
|
||||
mock_run.return_value = _mock_claude_success({
|
||||
"result": learner_output,
|
||||
"cost_usd": 0.042,
|
||||
"usage": {"total_tokens": 3000},
|
||||
})
|
||||
|
||||
step_results = [{"role": "debugger", "raw_output": "output"}]
|
||||
_run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
costs = models.get_cost_summary(conn, days=1)
|
||||
assert len(costs) == 1
|
||||
assert costs[0]["project_id"] == "vdol"
|
||||
assert costs[0]["total_cost_usd"] == pytest.approx(0.042)
|
||||
assert costs[0]["total_tokens"] == 3000
|
||||
|
||||
# -----------------------------------------------------------------------
|
||||
# KIN-061: Regression — валидация поля type в decision
|
||||
# -----------------------------------------------------------------------
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_valid_type_gotcha_is_saved_as_is(self, mock_run, conn):
|
||||
"""KIN-061: валидный тип 'gotcha' сохраняется без изменений."""
|
||||
learner_output = json.dumps({
|
||||
"decisions": [
|
||||
{"type": "gotcha", "title": "Use WAL mode", "description": "Concurrent reads need WAL", "tags": []},
|
||||
]
|
||||
})
|
||||
mock_run.return_value = _mock_claude_success({"result": learner_output})
|
||||
|
||||
result = _run_learning_extraction(conn, "VDOL-001", "vdol", [{"role": "debugger", "raw_output": "x"}])
|
||||
|
||||
assert result["added"] == 1
|
||||
d = conn.execute("SELECT type FROM decisions WHERE project_id='vdol'").fetchone()
|
||||
assert d["type"] == "gotcha"
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_invalid_type_falls_back_to_decision(self, mock_run, conn):
|
||||
"""KIN-061: невалидный тип 'unknown_type' заменяется на 'decision'."""
|
||||
learner_output = json.dumps({
|
||||
"decisions": [
|
||||
{"type": "unknown_type", "title": "Some title", "description": "Some desc", "tags": []},
|
||||
]
|
||||
})
|
||||
mock_run.return_value = _mock_claude_success({"result": learner_output})
|
||||
|
||||
result = _run_learning_extraction(conn, "VDOL-001", "vdol", [{"role": "debugger", "raw_output": "x"}])
|
||||
|
||||
assert result["added"] == 1
|
||||
d = conn.execute("SELECT type FROM decisions WHERE project_id='vdol'").fetchone()
|
||||
assert d["type"] == "decision"
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_missing_type_falls_back_to_decision(self, mock_run, conn):
|
||||
"""KIN-061: отсутствующий ключ 'type' в decision заменяется на 'decision'."""
|
||||
learner_output = json.dumps({
|
||||
"decisions": [
|
||||
{"title": "No type key here", "description": "desc without type", "tags": []},
|
||||
]
|
||||
})
|
||||
mock_run.return_value = _mock_claude_success({"result": learner_output})
|
||||
|
||||
result = _run_learning_extraction(conn, "VDOL-001", "vdol", [{"role": "debugger", "raw_output": "x"}])
|
||||
|
||||
assert result["added"] == 1
|
||||
d = conn.execute("SELECT type FROM decisions WHERE project_id='vdol'").fetchone()
|
||||
assert d["type"] == "decision"
|
||||
|
||||
|
||||
# -----------------------------------------------------------------------
|
||||
# KIN-062: KIN_LEARNER_TIMEOUT — отдельный таймаут для learner-агента
|
||||
# -----------------------------------------------------------------------
|
||||
|
||||
@patch.dict("os.environ", {"KIN_LEARNER_TIMEOUT": ""}, clear=False)
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_learner_uses_120s_default_timeout(self, mock_run, conn):
|
||||
"""KIN-062: по умолчанию learner использует таймаут 120s (KIN_LEARNER_TIMEOUT не задан)."""
|
||||
mock_run.return_value = _mock_claude_success({"result": json.dumps({"decisions": []})})
|
||||
|
||||
step_results = [{"role": "debugger", "raw_output": "output"}]
|
||||
_run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
call_kwargs = mock_run.call_args[1]
|
||||
assert call_kwargs.get("timeout") == 120
|
||||
|
||||
@patch.dict("os.environ", {"KIN_LEARNER_TIMEOUT": "300"}, clear=False)
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_learner_uses_custom_timeout_from_env(self, mock_run, conn):
|
||||
"""KIN-062: KIN_LEARNER_TIMEOUT переопределяет дефолтный таймаут learner-агента."""
|
||||
mock_run.return_value = _mock_claude_success({"result": json.dumps({"decisions": []})})
|
||||
|
||||
step_results = [{"role": "debugger", "raw_output": "output"}]
|
||||
_run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
call_kwargs = mock_run.call_args[1]
|
||||
assert call_kwargs.get("timeout") == 300
|
||||
|
||||
@patch.dict("os.environ", {"KIN_LEARNER_TIMEOUT": "60", "KIN_AGENT_TIMEOUT": "900"}, clear=False)
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_learner_timeout_independent_of_agent_timeout(self, mock_run, conn):
|
||||
"""KIN-062: KIN_LEARNER_TIMEOUT не зависит от KIN_AGENT_TIMEOUT."""
|
||||
mock_run.return_value = _mock_claude_success({"result": json.dumps({"decisions": []})})
|
||||
|
||||
step_results = [{"role": "debugger", "raw_output": "output"}]
|
||||
_run_learning_extraction(conn, "VDOL-001", "vdol", step_results)
|
||||
|
||||
call_kwargs = mock_run.call_args[1]
|
||||
assert call_kwargs.get("timeout") == 60
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# KIN-056: Regression — web path timeout parity with CLI
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
class TestRegressionKIN056:
|
||||
"""Регрессионные тесты KIN-056: агенты таймаутили через 300s из web, но не из CLI.
|
||||
|
||||
Причина: noninteractive режим использовал timeout=300s.
|
||||
Web API всегда устанавливает KIN_NONINTERACTIVE=1, поэтому таймаут был 300s.
|
||||
Фикс: единый timeout=600s независимо от noninteractive (переопределяется KIN_AGENT_TIMEOUT).
|
||||
|
||||
Каждый тест ПАДАЛ бы со старым кодом (timeout=300 для noninteractive)
|
||||
и ПРОХОДИТ после фикса.
|
||||
"""
|
||||
|
||||
@patch.dict("os.environ", {"KIN_NONINTERACTIVE": "1", "KIN_AGENT_TIMEOUT": ""})
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_web_noninteractive_env_does_not_use_300s(self, mock_run, conn):
|
||||
"""Web путь устанавливает KIN_NONINTERACTIVE=1. До фикса это давало timeout=300s."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
run_agent(conn, "debugger", "VDOL-001", "vdol")
|
||||
call_kwargs = mock_run.call_args[1]
|
||||
assert call_kwargs.get("timeout") != 300, (
|
||||
"Регрессия KIN-056: timeout не должен быть 300s в noninteractive режиме"
|
||||
)
|
||||
|
||||
@patch.dict("os.environ", {"KIN_NONINTERACTIVE": "1", "KIN_AGENT_TIMEOUT": ""})
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_web_noninteractive_timeout_is_600(self, mock_run, conn):
|
||||
"""Web путь: KIN_NONINTERACTIVE=1 → timeout = 600s (не 300s)."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
run_agent(conn, "debugger", "VDOL-001", "vdol")
|
||||
call_kwargs = mock_run.call_args[1]
|
||||
assert call_kwargs.get("timeout") == 600
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_web_and_cli_paths_use_same_timeout(self, mock_run, conn):
|
||||
"""Таймаут через web-путь (KIN_NONINTERACTIVE=1) == таймаут CLI (noninteractive=True)."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
|
||||
# Web path: env var KIN_NONINTERACTIVE=1, noninteractive param not set
|
||||
with patch.dict("os.environ", {"KIN_NONINTERACTIVE": "1", "KIN_AGENT_TIMEOUT": ""}):
|
||||
run_agent(conn, "debugger", "VDOL-001", "vdol", noninteractive=False)
|
||||
web_timeout = mock_run.call_args[1].get("timeout")
|
||||
|
||||
mock_run.reset_mock()
|
||||
|
||||
# CLI path: noninteractive=True, no env var
|
||||
with patch.dict("os.environ", {"KIN_NONINTERACTIVE": "", "KIN_AGENT_TIMEOUT": ""}):
|
||||
run_agent(conn, "debugger", "VDOL-001", "vdol", noninteractive=True)
|
||||
cli_timeout = mock_run.call_args[1].get("timeout")
|
||||
|
||||
assert web_timeout == cli_timeout, (
|
||||
f"Таймаут web ({web_timeout}s) != CLI ({cli_timeout}s) — регрессия KIN-056"
|
||||
)
|
||||
|
||||
@patch.dict("os.environ", {"KIN_NONINTERACTIVE": "1", "KIN_AGENT_TIMEOUT": "900"})
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_web_noninteractive_respects_kin_agent_timeout_override(self, mock_run, conn):
|
||||
"""Web путь: KIN_AGENT_TIMEOUT переопределяет дефолтный таймаут даже при KIN_NONINTERACTIVE=1."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
run_agent(conn, "debugger", "VDOL-001", "vdol")
|
||||
call_kwargs = mock_run.call_args[1]
|
||||
assert call_kwargs.get("timeout") == 900
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# KIN-057: claude CLI в PATH при запуске через launchctl
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
class TestClaudePath:
|
||||
"""Регрессионные тесты KIN-057: launchctl-демоны могут не видеть claude в PATH."""
|
||||
|
||||
def test_build_claude_env_contains_extra_paths(self):
|
||||
"""_build_claude_env должен добавить /opt/homebrew/bin и /usr/local/bin в PATH."""
|
||||
env = _build_claude_env()
|
||||
path_dirs = env["PATH"].split(":")
|
||||
for extra_dir in _EXTRA_PATH_DIRS:
|
||||
assert extra_dir in path_dirs, (
|
||||
f"Регрессия KIN-057: {extra_dir} не найден в PATH, сгенерированном _build_claude_env"
|
||||
)
|
||||
|
||||
def test_build_claude_env_no_duplicate_paths(self):
|
||||
"""_build_claude_env не должен дублировать уже существующие пути."""
|
||||
env = _build_claude_env()
|
||||
path_dirs = env["PATH"].split(":")
|
||||
seen = set()
|
||||
for d in path_dirs:
|
||||
assert d not in seen, f"Дублирующийся PATH entry: {d}"
|
||||
seen.add(d)
|
||||
|
||||
def test_build_claude_env_preserves_existing_path(self):
|
||||
"""_build_claude_env должен сохранять уже существующие пути."""
|
||||
with patch.dict("os.environ", {"PATH": "/custom/bin:/usr/bin:/bin"}):
|
||||
env = _build_claude_env()
|
||||
path_dirs = env["PATH"].split(":")
|
||||
assert "/custom/bin" in path_dirs
|
||||
assert "/usr/bin" in path_dirs
|
||||
|
||||
def test_resolve_claude_cmd_returns_string(self):
|
||||
"""_resolve_claude_cmd должен всегда возвращать строку."""
|
||||
cmd = _resolve_claude_cmd()
|
||||
assert isinstance(cmd, str)
|
||||
assert len(cmd) > 0
|
||||
|
||||
def test_resolve_claude_cmd_fallback_when_not_found(self):
|
||||
"""_resolve_claude_cmd должен вернуть 'claude' если CLI не найден в PATH."""
|
||||
with patch("agents.runner.shutil.which", return_value=None):
|
||||
cmd = _resolve_claude_cmd()
|
||||
assert cmd == "claude"
|
||||
|
||||
def test_resolve_claude_cmd_returns_full_path_when_found(self):
|
||||
"""_resolve_claude_cmd должен вернуть полный путь если claude найден."""
|
||||
with patch("agents.runner.shutil.which", return_value="/opt/homebrew/bin/claude"):
|
||||
cmd = _resolve_claude_cmd()
|
||||
assert cmd == "/opt/homebrew/bin/claude"
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_run_claude_passes_env_to_subprocess(self, mock_run, conn):
|
||||
"""_run_claude должен передавать env= в subprocess.run (а не наследовать голый PATH)."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
run_agent(conn, "debugger", "VDOL-001", "vdol")
|
||||
call_kwargs = mock_run.call_args[1]
|
||||
assert "env" in call_kwargs, (
|
||||
"Регрессия KIN-057: subprocess.run должен получать явный env с расширенным PATH"
|
||||
)
|
||||
assert call_kwargs["env"] is not None
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_run_claude_env_has_homebrew_in_path(self, mock_run, conn):
|
||||
"""env переданный в subprocess.run должен содержать /opt/homebrew/bin в PATH."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
run_agent(conn, "debugger", "VDOL-001", "vdol")
|
||||
call_kwargs = mock_run.call_args[1]
|
||||
env = call_kwargs.get("env", {})
|
||||
assert "/opt/homebrew/bin" in env.get("PATH", ""), (
|
||||
"Регрессия KIN-057: /opt/homebrew/bin не найден в env['PATH'] subprocess.run"
|
||||
)
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_file_not_found_returns_127(self, mock_run, conn):
|
||||
"""Если claude не найден (FileNotFoundError), должен вернуться returncode 127."""
|
||||
mock_run.side_effect = FileNotFoundError("claude not found")
|
||||
result = run_agent(conn, "debugger", "VDOL-001", "vdol")
|
||||
assert result["success"] is False
|
||||
assert "not found" in (result.get("error") or "").lower()
|
||||
|
||||
@patch.dict("os.environ", {"PATH": ""})
|
||||
def test_launchctl_empty_path_build_env_adds_extra_dirs(self):
|
||||
"""Регрессия KIN-057: когда launchctl запускает с пустым PATH,
|
||||
_build_claude_env должен добавить _EXTRA_PATH_DIRS чтобы claude был доступен.
|
||||
Без фикса: os.environ["PATH"]="" → shutil.which("claude") → None → FileNotFoundError.
|
||||
После фикса: _build_claude_env строит PATH с /opt/homebrew/bin и др.
|
||||
"""
|
||||
env = _build_claude_env()
|
||||
path_dirs = env["PATH"].split(":")
|
||||
# Явная проверка каждой критичной директории
|
||||
for extra_dir in _EXTRA_PATH_DIRS:
|
||||
assert extra_dir in path_dirs, (
|
||||
f"KIN-057: при пустом os PATH директория {extra_dir} должна быть добавлена"
|
||||
)
|
||||
|
||||
@patch.dict("os.environ", {"PATH": ""})
|
||||
def test_launchctl_empty_path_shutil_which_fails_without_fix(self):
|
||||
"""Воспроизводит сломанное поведение: при PATH='' shutil.which возвращает None.
|
||||
Это точно то, что происходило до фикса — launchctl не видел claude.
|
||||
Тест документирует, ПОЧЕМУ нужен _build_claude_env вместо прямого os.environ.
|
||||
"""
|
||||
import shutil
|
||||
# Без фикса: поиск с пустым PATH не найдёт claude
|
||||
result_without_fix = shutil.which("claude", path="")
|
||||
assert result_without_fix is None, (
|
||||
"Если этот assert упал — shutil.which нашёл claude в пустом PATH, "
|
||||
"что невозможно. Ожидаем None — именно поэтому нужен _build_claude_env."
|
||||
)
|
||||
# С фиксом: _resolve_claude_cmd строит расширенный PATH и находит claude
|
||||
# (или возвращает fallback "claude", но не бросает FileNotFoundError)
|
||||
cmd = _resolve_claude_cmd()
|
||||
assert isinstance(cmd, str) and len(cmd) > 0, (
|
||||
"KIN-057: _resolve_claude_cmd должен возвращать строку даже при пустом os PATH"
|
||||
)
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# KIN-063: TestCompletionMode — auto_complete + last-step role check
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
class TestCompletionMode:
|
||||
"""auto_complete mode срабатывает только если последний шаг — tester или reviewer."""
|
||||
|
||||
@patch("core.followup.generate_followups")
|
||||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_auto_complete_with_tester_last_sets_done(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""auto_complete + последний шаг tester → status=done (Decision #29)."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "developer", "brief": "fix"}, {"role": "tester", "brief": "test"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
task = models.get_task(conn, "VDOL-001")
|
||||
assert task["status"] == "done"
|
||||
|
||||
@patch("core.followup.generate_followups")
|
||||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_auto_complete_with_reviewer_last_sets_done(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""auto_complete + последний шаг reviewer → status=done."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "developer", "brief": "fix"}, {"role": "reviewer", "brief": "review"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
task = models.get_task(conn, "VDOL-001")
|
||||
assert task["status"] == "done"
|
||||
|
||||
@patch("core.followup.generate_followups")
|
||||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_auto_complete_without_tester_last_sets_review(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""auto_complete + последний шаг НЕ tester/reviewer → status=review (Decision #29)."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
models.update_project(conn, "vdol", execution_mode="auto_complete")
|
||||
steps = [{"role": "developer", "brief": "fix"}, {"role": "debugger", "brief": "debug"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
task = models.get_task(conn, "VDOL-001")
|
||||
assert task["status"] == "review", (
|
||||
"Регрессия KIN-063: auto_complete без tester/reviewer последним НЕ должен авто-завершать"
|
||||
)
|
||||
|
||||
@patch("core.followup.generate_followups")
|
||||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_legacy_auto_mode_value_not_recognized(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""Регрессия: старое значение 'auto' больше не является валидным режимом.
|
||||
|
||||
После KIN-063 'auto' → 'auto_complete'. Если в DB осталось 'auto' (без миграции),
|
||||
runner НЕ должен авто-завершать — это 'review'-ветка (безопасный fallback).
|
||||
(Decision #29)
|
||||
"""
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
# Прямой SQL-апдейт, обходя validate_completion_mode, чтобы симулировать
|
||||
# старую запись в БД без миграции
|
||||
conn.execute("UPDATE projects SET execution_mode='auto' WHERE id='vdol'")
|
||||
conn.commit()
|
||||
steps = [{"role": "developer", "brief": "fix"}, {"role": "tester", "brief": "test"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
task = models.get_task(conn, "VDOL-001")
|
||||
assert task["status"] == "review", (
|
||||
"Регрессия: 'auto' (старый формат) не должен срабатывать как auto_complete"
|
||||
)
|
||||
|
||||
@patch("core.followup.generate_followups")
|
||||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_review_mode_with_tester_last_keeps_task_in_review(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""review mode + последний шаг tester → task.status == 'review', НЕ done (ждёт ручного approve)."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "all tests pass"})
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
# Проект и задача остаются в дефолтном 'review' mode
|
||||
steps = [{"role": "developer", "brief": "fix"}, {"role": "tester", "brief": "test"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
task = models.get_task(conn, "VDOL-001")
|
||||
assert task["status"] == "review"
|
||||
assert task["status"] != "done", (
|
||||
"KIN-063: review mode не должен авто-завершать задачу даже если tester последний"
|
||||
)
|
||||
|
||||
@patch("core.followup.generate_followups")
|
||||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_project_review_overrides_no_task_completion_mode(self, mock_run, mock_hooks, mock_followup, conn):
|
||||
"""Project execution_mode='review' + задача без override → pipeline завершается в 'review'.
|
||||
|
||||
Сценарий: PM выбрал auto_complete, но проект настроен на 'review' (ручной override человека).
|
||||
Задача не имеет task-level execution_mode, поэтому get_effective_mode возвращает project-level 'review'.
|
||||
"""
|
||||
mock_run.return_value = _mock_claude_success({"result": "ok"})
|
||||
mock_hooks.return_value = []
|
||||
mock_followup.return_value = {"created": [], "pending_actions": []}
|
||||
|
||||
# Проект явно в 'review', задача без execution_mode
|
||||
models.update_project(conn, "vdol", execution_mode="review")
|
||||
# task VDOL-001 создана без execution_mode (None) — fixture
|
||||
steps = [{"role": "developer", "brief": "fix"}, {"role": "tester", "brief": "test"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
assert result["mode"] == "review"
|
||||
task = models.get_task(conn, "VDOL-001")
|
||||
assert task["status"] == "review", (
|
||||
"KIN-063: project-level 'review' должен применяться когда задача не имеет override"
|
||||
)
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# KIN-048: _run_autocommit — флаг, git path, env=
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
class TestAutocommit:
|
||||
"""KIN-048: _run_autocommit — autocommit_enabled флаг, shutil.which, env= regression."""
|
||||
|
||||
def test_disabled_project_skips_subprocess(self, conn):
|
||||
"""autocommit_enabled=0 (дефолт) → subprocess не вызывается."""
|
||||
with patch("agents.runner.subprocess.run") as mock_run:
|
||||
_run_autocommit(conn, "VDOL-001", "vdol")
|
||||
mock_run.assert_not_called()
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
@patch("agents.runner.shutil.which")
|
||||
def test_enabled_calls_git_add_and_commit(self, mock_which, mock_run, conn, tmp_path):
|
||||
"""autocommit_enabled=1 → вызываются git add -A и git commit с task_id и title."""
|
||||
mock_which.return_value = "/usr/bin/git"
|
||||
mock_run.return_value = MagicMock(returncode=0)
|
||||
models.update_project(conn, "vdol", autocommit_enabled=1, path=str(tmp_path))
|
||||
|
||||
_run_autocommit(conn, "VDOL-001", "vdol")
|
||||
|
||||
assert mock_run.call_count == 2
|
||||
add_cmd = mock_run.call_args_list[0][0][0]
|
||||
assert add_cmd == ["/usr/bin/git", "add", "-A"]
|
||||
commit_cmd = mock_run.call_args_list[1][0][0]
|
||||
assert commit_cmd[0] == "/usr/bin/git"
|
||||
assert commit_cmd[1] == "commit"
|
||||
assert "VDOL-001" in commit_cmd[-1]
|
||||
assert "Fix bug" in commit_cmd[-1]
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_nothing_to_commit_no_exception(self, mock_run, conn, tmp_path):
|
||||
"""returncode=1 (nothing to commit) → исключение не бросается."""
|
||||
mock_run.return_value = MagicMock(returncode=1)
|
||||
models.update_project(conn, "vdol", autocommit_enabled=1, path=str(tmp_path))
|
||||
|
||||
_run_autocommit(conn, "VDOL-001", "vdol") # must not raise
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_passes_env_to_subprocess(self, mock_run, conn, tmp_path):
|
||||
"""Regression #33: env= должен передаваться в subprocess.run."""
|
||||
mock_run.return_value = MagicMock(returncode=0)
|
||||
models.update_project(conn, "vdol", autocommit_enabled=1, path=str(tmp_path))
|
||||
|
||||
_run_autocommit(conn, "VDOL-001", "vdol")
|
||||
|
||||
for call in mock_run.call_args_list:
|
||||
kwargs = call[1]
|
||||
assert "env" in kwargs, "Regression #33: subprocess.run должен получать env="
|
||||
assert "/opt/homebrew/bin" in kwargs["env"].get("PATH", "")
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
@patch("agents.runner.shutil.which")
|
||||
def test_resolves_git_via_shutil_which(self, mock_which, mock_run, conn, tmp_path):
|
||||
"""Regression #32: git резолвится через shutil.which, а не hardcoded 'git'."""
|
||||
mock_which.return_value = "/opt/homebrew/bin/git"
|
||||
mock_run.return_value = MagicMock(returncode=0)
|
||||
models.update_project(conn, "vdol", autocommit_enabled=1, path=str(tmp_path))
|
||||
|
||||
_run_autocommit(conn, "VDOL-001", "vdol")
|
||||
|
||||
git_which_calls = [c for c in mock_which.call_args_list if c[0][0] == "git"]
|
||||
assert len(git_which_calls) > 0, "Regression #32: shutil.which должен вызываться для git"
|
||||
first_cmd = mock_run.call_args_list[0][0][0]
|
||||
assert first_cmd[0] == "/opt/homebrew/bin/git"
|
||||
|
||||
@patch("agents.runner.subprocess.run")
|
||||
@patch("agents.runner.shutil.which")
|
||||
def test_git_not_found_no_crash_logs_warning(self, mock_which, mock_run, conn, tmp_path):
|
||||
"""shutil.which(git) → None → fallback 'git' → FileNotFoundError → no crash, WARNING logged."""
|
||||
mock_which.return_value = None # git не найден в PATH
|
||||
mock_run.side_effect = FileNotFoundError("git: command not found")
|
||||
models.update_project(conn, "vdol", autocommit_enabled=1, path=str(tmp_path))
|
||||
|
||||
with patch("agents.runner._logger") as mock_logger:
|
||||
_run_autocommit(conn, "VDOL-001", "vdol") # не должен бросать исключение
|
||||
|
||||
mock_logger.warning.assert_called_once()
|
||||
|
||||
@patch("agents.runner._run_autocommit")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_autocommit_not_called_on_failed_pipeline(self, mock_run, mock_autocommit, conn):
|
||||
"""Pipeline failure → _run_autocommit must NOT be called (gotcha #41)."""
|
||||
mock_run.return_value = _mock_claude_failure("compilation error")
|
||||
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is False
|
||||
mock_autocommit.assert_not_called()
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# KIN-055: execution_mode='review' при переводе задачи в статус review
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
class TestReviewModeExecutionMode:
|
||||
"""Регрессия KIN-055: execution_mode должен быть 'review', а не NULL после pipeline в review mode."""
|
||||
|
||||
def test_task_execution_mode_is_null_before_pipeline(self, conn):
|
||||
"""Граничный случай: execution_mode IS NULL до запуска pipeline (задача только создана)."""
|
||||
task = models.get_task(conn, "VDOL-001")
|
||||
assert task["execution_mode"] is None, (
|
||||
"Задача должна иметь NULL execution_mode до выполнения pipeline"
|
||||
)
|
||||
|
||||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_review_mode_sets_execution_mode_review(self, mock_run, mock_hooks, conn):
|
||||
"""После pipeline в review mode task.execution_mode должно быть 'review', а не NULL."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "done"})
|
||||
mock_hooks.return_value = []
|
||||
|
||||
steps = [{"role": "debugger", "brief": "find bug"}]
|
||||
result = run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
assert result["success"] is True
|
||||
task = models.get_task(conn, "VDOL-001")
|
||||
assert task["status"] == "review"
|
||||
# Регрессионная проверка KIN-055: execution_mode не должен быть NULL
|
||||
assert task["execution_mode"] is not None, (
|
||||
"Регрессия KIN-055: execution_mode не должен быть NULL после перевода задачи в статус review"
|
||||
)
|
||||
assert task["execution_mode"] == "review"
|
||||
|
||||
@patch("agents.runner.run_hooks")
|
||||
@patch("agents.runner.subprocess.run")
|
||||
def test_review_mode_execution_mode_persisted_in_db(self, mock_run, mock_hooks, conn):
|
||||
"""execution_mode='review' должно сохраняться в SQLite напрямую, минуя ORM-слой."""
|
||||
mock_run.return_value = _mock_claude_success({"result": "done"})
|
||||
mock_hooks.return_value = []
|
||||
|
||||
steps = [{"role": "debugger", "brief": "find"}]
|
||||
run_pipeline(conn, "VDOL-001", steps)
|
||||
|
||||
row = conn.execute(
|
||||
"SELECT execution_mode FROM tasks WHERE id='VDOL-001'"
|
||||
).fetchone()
|
||||
assert row is not None
|
||||
assert row["execution_mode"] == "review", (
|
||||
"Регрессия KIN-055: execution_mode должен быть 'review' в SQLite после pipeline"
|
||||
)
|
||||
|
|
|
|||
62
web/api.py
62
web/api.py
|
|
@ -3,6 +3,8 @@ Kin Web API — FastAPI backend reading ~/.kin/kin.db via core.models.
|
|||
Run: uvicorn web.api:app --reload --port 8420
|
||||
"""
|
||||
|
||||
import logging
|
||||
import shutil
|
||||
import subprocess
|
||||
import sys
|
||||
from pathlib import Path
|
||||
|
|
@ -18,6 +20,7 @@ from pydantic import BaseModel
|
|||
|
||||
from core.db import init_db
|
||||
from core import models
|
||||
from core.models import VALID_COMPLETION_MODES
|
||||
from agents.bootstrap import (
|
||||
detect_tech_stack, detect_modules, extract_decisions_from_claude_md,
|
||||
find_vault_root, scan_obsidian, save_to_db,
|
||||
|
|
@ -25,6 +28,62 @@ from agents.bootstrap import (
|
|||
|
||||
DB_PATH = Path.home() / ".kin" / "kin.db"
|
||||
|
||||
_logger = logging.getLogger("kin")
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Startup: verify claude CLI is available in PATH
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
def _check_claude_available() -> None:
|
||||
"""Warn at startup if claude CLI cannot be found in PATH.
|
||||
|
||||
launchctl daemons run with a stripped environment and may not see
|
||||
/opt/homebrew/bin where claude is typically installed.
|
||||
See Decision #28.
|
||||
"""
|
||||
from agents.runner import _build_claude_env # avoid circular import at module level
|
||||
env = _build_claude_env()
|
||||
claude_path = shutil.which("claude", path=env["PATH"])
|
||||
if claude_path:
|
||||
_logger.info("claude CLI found: %s", claude_path)
|
||||
else:
|
||||
_logger.warning(
|
||||
"WARNING: claude CLI not found in PATH (%s). "
|
||||
"Agent pipelines will fail with returncode 127. "
|
||||
"Fix: add /opt/homebrew/bin to EnvironmentVariables.PATH in "
|
||||
"~/Library/LaunchAgents/com.kin.api.plist and reload with: "
|
||||
"launchctl unload ~/Library/LaunchAgents/com.kin.api.plist && "
|
||||
"launchctl load ~/Library/LaunchAgents/com.kin.api.plist",
|
||||
env.get("PATH", ""),
|
||||
)
|
||||
|
||||
|
||||
def _check_git_available() -> None:
|
||||
"""Warn at startup if git cannot be found in PATH.
|
||||
|
||||
launchctl daemons run with a stripped environment and may not see
|
||||
git in the standard directories. See Decision #28.
|
||||
"""
|
||||
from agents.runner import _build_claude_env # avoid circular import at module level
|
||||
env = _build_claude_env()
|
||||
git_path = shutil.which("git", path=env["PATH"])
|
||||
if git_path:
|
||||
_logger.info("git found: %s", git_path)
|
||||
else:
|
||||
_logger.warning(
|
||||
"WARNING: git not found in PATH (%s). "
|
||||
"Autocommit will fail silently. "
|
||||
"Fix: add git directory to EnvironmentVariables.PATH in "
|
||||
"~/Library/LaunchAgents/com.kin.api.plist and reload with: "
|
||||
"launchctl unload ~/Library/LaunchAgents/com.kin.api.plist && "
|
||||
"launchctl load ~/Library/LaunchAgents/com.kin.api.plist",
|
||||
env.get("PATH", ""),
|
||||
)
|
||||
|
||||
|
||||
_check_claude_available()
|
||||
_check_git_available()
|
||||
|
||||
app = FastAPI(title="Kin API", version="0.1.0")
|
||||
|
||||
app.add_middleware(
|
||||
|
|
@ -162,7 +221,7 @@ class TaskPatch(BaseModel):
|
|||
|
||||
|
||||
VALID_STATUSES = set(models.VALID_TASK_STATUSES)
|
||||
VALID_EXECUTION_MODES = {"auto", "review"}
|
||||
VALID_EXECUTION_MODES = VALID_COMPLETION_MODES
|
||||
|
||||
|
||||
@app.patch("/api/tasks/{task_id}")
|
||||
|
|
@ -361,6 +420,7 @@ def run_task(task_id: str):
|
|||
cmd,
|
||||
cwd=str(kin_root),
|
||||
stdout=subprocess.DEVNULL,
|
||||
stderr=subprocess.DEVNULL,
|
||||
stdin=subprocess.DEVNULL,
|
||||
env=env,
|
||||
)
|
||||
|
|
|
|||
Loading…
Add table
Add a link
Reference in a new issue