Files
orchestrator/tests/test_auto_approve_brd.py
claude-bot a6d0ba51c0 feat(labels): auto-mode by Plane labels — autoApprove + autoDeploy (ORCH-089)
Lift the two HUMAN gates that block an autonomous batch run (epic ORCH-088):
the BRD gate (analysis: manual Approved) and the prod-deploy gate (deploy
Phase A: manual Confirm Deploy, ORCH-059). Selective (a Plane label on the
issue), declarative, reversible, and WITHOUT touching a single technical check.

Additive, mirroring the conditional sub-gates (ORCH-035/043/058/088): leaf
src/labels.py (never-raise) + two point insertions + config flags.
STAGE_TRANSITIONS / QG_CHECKS / check_* / DB schema are NOT touched.

- autoApprove: врезка in _handle_analysis_approved_flow (files_ok branch) ->
  set_issue_approved + log/Telegram/Plane-comment + advance_stage(
  finished_agent=None) — the SAME path a human Approved takes (approved-via-
  status -> analysis->architecture + mark_brd_review_ended). No duplicated
  transition logic; re-entrancy safe.
- autoDeploy: врезка in _handle_self_deploy_phase_a after advance to deploy +
  clear_state -> log/Telegram/Plane-comment + _handle_self_deploy_phase_b
  (INITIATED marker, Deploying, finalizer). Only the indicative human steps are
  skipped. BR-5 holds structurally: Phase A is reached only after the green edge
  sub-gates, so autoDeploy can never deploy a broken build.
- plane_sync: fetch_issue_labels (None on error != []), get_project_labels
  ({normalized_name->uuid}, TTL cache, ambiguity sentinel), set_issue_approved.
- config flags: auto_label_enabled (kill-switch), auto_approve_label/
  auto_deploy_label, auto_label_repos (empty -> self-hosting only),
  auto_label_states_ttl_s. applies() (local) checked FIRST; has_label (network)
  only when applies==True -> zero network / zero regression when disabled (AC-8).
- Fail-safe (never auto on doubt), transparency via log+Telegram+Plane+card,
  read-only auto_labels block in GET /queue.
- Tests TC-01..TC-26 across 7 modules; docs (CLAUDE.md, architecture README,
  CHANGELOG) updated in the same PR.

Refs: ORCH-089

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
2026-06-09 12:31:24 +03:00

191 lines
7.6 KiB
Python

"""ORCH-089 — autoApprove врезка in _handle_analysis_approved_flow.
Covers (04-test-plan.yaml):
TC-10 autoApprove + artifacts ready -> auto-advance analysis->architecture,
Approved set, brd_review_ended clock closed.
TC-11 no autoApprove label -> prior behaviour: In Review, return w/o advance.
TC-12 autoApprove but artifacts missing (check_analysis_complete FAIL) -> NO
advance (AC-5 for BRD).
TC-13 autoApprove goes through the SAME advance path as a manual Approved (no
duplicated transition logic; idempotent — stage lands on architecture).
TC-14 autoApprove logged + Telegram + Plane comment (transparency AC-7).
"""
import os
import tempfile
import pytest
_test_db = os.path.join(tempfile.gettempdir(), "test_auto_approve_brd.db")
os.environ["ORCH_DB_PATH"] = _test_db
os.environ["ORCH_REPOS_DIR"] = tempfile.gettempdir()
os.environ.setdefault("ORCH_GITEA_TOKEN", "test-token")
os.environ.setdefault("ORCH_PLANE_API_TOKEN", "test-token")
from unittest.mock import MagicMock # noqa: E402
import src.db as _db # noqa: E402
from src.db import init_db, get_db # noqa: E402
from src import stage_engine # noqa: E402
from src import labels # noqa: E402
from src.stage_engine import advance_stage # noqa: E402
def _files_ok(*a, **k):
return (True, "ok")
def _files_fail(*a, **k):
return (False, "missing artifacts")
@pytest.fixture(autouse=True)
def fresh(monkeypatch, tmp_path):
monkeypatch.setattr(_db.settings, "db_path", _test_db)
if os.path.exists(_test_db):
os.unlink(_test_db)
init_db()
# Silence Plane/Telegram side effects; capture the transparency channels.
for name in ("notify_stage_change", "notify_qg_failure", "plane_notify_stage",
"plane_notify_qg", "set_issue_in_review", "set_issue_needs_input",
"set_issue_approved", "notify_approve_requested"):
monkeypatch.setattr(stage_engine, name, MagicMock(), raising=False)
# Avoid worktree access in the analyst "ready" comment builder.
monkeypatch.setattr(stage_engine, "_build_analyst_ready_comment",
lambda *a, **k: "ready", raising=False)
monkeypatch.setattr(stage_engine, "plane_add_comment", MagicMock())
monkeypatch.setattr(stage_engine, "send_telegram", MagicMock())
yield
def _make_task(stage="analysis", repo="orchestrator", branch="feature/ORCH-089-x",
wi="ORCH-089"):
conn = get_db()
cur = conn.execute(
"INSERT INTO tasks (plane_id, work_item_id, repo, branch, stage) "
"VALUES (?, ?, ?, ?, ?)",
(f"plane-{wi}", wi, repo, branch, stage),
)
tid = cur.lastrowid
conn.commit()
conn.close()
return tid
def _stage_of(task_id):
conn = get_db()
row = conn.execute("SELECT stage FROM tasks WHERE id=?", (task_id,)).fetchone()
conn.close()
return row[0]
def _brd_ended(task_id):
conn = get_db()
row = conn.execute(
"SELECT brd_review_ended_at FROM tasks WHERE id=?", (task_id,)
).fetchone()
conn.close()
return row[0]
def _patch_complete_gate(monkeypatch, ok=True):
gate = _files_ok if ok else _files_fail
monkeypatch.setattr(
stage_engine, "QG_CHECKS",
{**stage_engine.QG_CHECKS, "check_analysis_complete": gate},
)
def _label(monkeypatch, present=True, applies=True):
monkeypatch.setattr(labels, "auto_approve_applies", lambda repo: applies)
monkeypatch.setattr(labels, "has_label", lambda w, name, p=None: present)
# --- TC-10 -----------------------------------------------------------------
def test_tc10_auto_approve_advances(monkeypatch):
_patch_complete_gate(monkeypatch, ok=True)
_label(monkeypatch, present=True)
tid = _make_task()
# The BRD-review clock was started when the task entered In Review; the
# advance closes it (mark_brd_review_ended only stamps when a start exists).
conn = get_db()
conn.execute(
"UPDATE tasks SET brd_review_started_at=datetime('now') WHERE id=?", (tid,)
)
conn.commit()
conn.close()
res = advance_stage(tid, "analysis", "orchestrator", "ORCH-089",
"feature/ORCH-089-x", finished_agent="analyst")
assert res.note == "auto-approved-via-label"
assert res.advanced is True
assert _stage_of(tid) == "architecture"
assert _brd_ended(tid) is not None # clock closed by mark_brd_review_ended
stage_engine.set_issue_approved.assert_called_once() # Approved indication
# --- TC-11 -----------------------------------------------------------------
def test_tc11_no_label_waits_for_human(monkeypatch):
_patch_complete_gate(monkeypatch, ok=True)
_label(monkeypatch, present=False)
tid = _make_task()
res = advance_stage(tid, "analysis", "orchestrator", "ORCH-089",
"feature/ORCH-089-x", finished_agent="analyst")
assert res.note == "analysis-in-review"
assert res.advanced is False
assert _stage_of(tid) == "analysis" # still waiting for a human
stage_engine.set_issue_in_review.assert_called_once()
stage_engine.set_issue_approved.assert_not_called()
# --- TC-12 -----------------------------------------------------------------
def test_tc12_missing_artifacts_no_auto(monkeypatch):
# autoApprove present, but artifacts incomplete -> files_ok False -> the
# autoApprove block (inside `if files_ok`) is never reached.
_patch_complete_gate(monkeypatch, ok=False)
_label(monkeypatch, present=True)
tid = _make_task()
res = advance_stage(tid, "analysis", "orchestrator", "ORCH-089",
"feature/ORCH-089-x", finished_agent="analyst")
assert res.advanced is False
assert _stage_of(tid) == "analysis"
assert res.note != "auto-approved-via-label"
stage_engine.set_issue_approved.assert_not_called()
# --- TC-13: same advance path / idempotent ---------------------------------
def test_tc13_same_advance_path_idempotent(monkeypatch):
_patch_complete_gate(monkeypatch, ok=True)
_label(monkeypatch, present=True)
tid = _make_task()
res = advance_stage(tid, "analysis", "orchestrator", "ORCH-089",
"feature/ORCH-089-x", finished_agent="analyst")
# The advance went through the unified path -> architect enqueued exactly once.
assert res.enqueued_agent == "architect"
conn = get_db()
n = conn.execute(
"SELECT COUNT(*) FROM jobs WHERE task_id=? AND agent='architect'", (tid,)
).fetchone()[0]
conn.close()
assert n == 1
# A later real Approved (webhook path, finished_agent=None) sees architecture,
# not analysis -> it cannot re-run the analysis advance (idempotent).
assert _stage_of(tid) == "architecture"
# --- TC-14: transparency ---------------------------------------------------
def test_tc14_transparency_channels(monkeypatch, caplog):
_patch_complete_gate(monkeypatch, ok=True)
_label(monkeypatch, present=True)
tid = _make_task()
import logging
with caplog.at_level(logging.INFO, logger="orchestrator.stage_engine"):
advance_stage(tid, "analysis", "orchestrator", "ORCH-089",
"feature/ORCH-089-x", finished_agent="analyst")
# (a) log mentions the label + auto-approve.
assert any("auto-approved" in r.message.lower() or "autoApprove" in r.message
for r in caplog.records)
# (b) Telegram fired; (c) a Plane comment authored by analyst about the auto-pass.
assert stage_engine.send_telegram.called
comment_calls = [c for c in stage_engine.plane_add_comment.call_args_list
if "авто-подтверждён" in c.args[1]]
assert comment_calls, "expected an auto-approve Plane comment"