Files
orchestrator/tests/test_auto_deploy.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

183 lines
7.8 KiB
Python

"""ORCH-089 — autoDeploy врезка in _handle_self_deploy_phase_a.
Covers (04-test-plan.yaml):
TC-15 autoDeploy + Phase A advance on `deploy` -> Phase B (initiate_deploy)
is auto-invoked.
TC-16 no autoDeploy label -> prior Phase A: Awaiting Deploy, wait for Confirm
Deploy.
TC-17 idempotent: INITIATED marker already present -> repeat auto-trigger no-op.
TC-18 non-self repo / out of scope -> no auto (Phase A/B only for self-hosting).
TC-19 autoDeploy logged + Telegram + Plane comment (transparency AC-7).
"""
import os
import tempfile
import pytest
_test_db = os.path.join(tempfile.gettempdir(), "test_auto_deploy.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 self_deploy # noqa: E402
from src import labels # noqa: E402
from src.stage_engine import advance_stage # noqa: E402
def _pass(*a, **k):
return (True, "ok")
@pytest.fixture(autouse=True)
def fresh_db(monkeypatch, tmp_path):
monkeypatch.setattr(_db.settings, "db_path", _test_db)
if os.path.exists(_test_db):
os.unlink(_test_db)
init_db()
monkeypatch.setattr(self_deploy.settings, "repos_dir", str(tmp_path))
monkeypatch.setattr(self_deploy.settings, "host_repos_dir", str(tmp_path))
monkeypatch.setattr(stage_engine.settings, "deploy_require_manual_approve", True)
# Pass all edge sub-gates so the deploy-staging -> deploy edge reaches Phase A.
monkeypatch.setattr(
stage_engine, "QG_CHECKS",
{**stage_engine.QG_CHECKS,
"check_staging_status": _pass,
"check_branch_mergeable": _pass,
"check_security_gate": _pass,
"check_staging_image_fresh": _pass},
)
# Default auto-mode flags ON (overridden per-test).
monkeypatch.setattr(stage_engine.settings, "auto_label_enabled", True, raising=False)
monkeypatch.setattr(stage_engine.settings, "auto_deploy_label", "autoDeploy", raising=False)
yield
@pytest.fixture(autouse=True)
def silence(monkeypatch):
for name in ("notify_stage_change", "notify_qg_failure", "plane_notify_stage",
"plane_notify_qg", "set_issue_in_review", "set_issue_awaiting_deploy",
"set_issue_deploying"):
monkeypatch.setattr(stage_engine, name, MagicMock(), raising=False)
monkeypatch.setattr(stage_engine, "plane_add_comment", MagicMock())
monkeypatch.setattr(stage_engine, "send_telegram", MagicMock())
def _make_task(stage="deploy-staging", 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 _label(monkeypatch, present=True, applies=True):
monkeypatch.setattr(labels, "auto_deploy_applies", lambda repo: applies)
monkeypatch.setattr(labels, "has_label", lambda w, name, p=None: present)
def _advance(tid, repo="orchestrator", wi="ORCH-089"):
return advance_stage(tid, "deploy-staging", repo, wi,
"feature/ORCH-089-x", finished_agent="deployer")
# --- TC-15 -----------------------------------------------------------------
def test_tc15_auto_deploy_initiates_phase_b(monkeypatch):
_label(monkeypatch, present=True)
initiate = MagicMock(return_value=(True, "ok"))
monkeypatch.setattr(stage_engine.self_deploy, "initiate_deploy", initiate)
tid = _make_task()
res = _advance(tid)
# Phase B ran via the same path a human Confirm Deploy takes.
initiate.assert_called_once()
assert res.note == "self-deploy-initiated"
assert self_deploy.has_marker("orchestrator", "ORCH-089", self_deploy.INITIATED)
# APPROVE_REQUESTED (the human ask) was SKIPPED on the auto path.
assert not self_deploy.has_marker(
"orchestrator", "ORCH-089", self_deploy.APPROVE_REQUESTED
)
# --- TC-16 -----------------------------------------------------------------
def test_tc16_no_label_waits_for_human(monkeypatch):
_label(monkeypatch, present=False)
initiate = MagicMock(return_value=(True, "ok"))
monkeypatch.setattr(stage_engine.self_deploy, "initiate_deploy", initiate)
tid = _make_task()
res = _advance(tid)
# Prior Phase A behaviour: approval-pending, no deploy initiated.
assert res.note == "self-deploy-approval-pending"
initiate.assert_not_called()
assert self_deploy.has_marker(
"orchestrator", "ORCH-089", self_deploy.APPROVE_REQUESTED
)
stage_engine.set_issue_awaiting_deploy.assert_called_once()
# --- TC-17: idempotency ----------------------------------------------------
def test_tc17_idempotent_initiated_marker(monkeypatch):
"""autoDeploy delegates prod-deploy to _handle_self_deploy_phase_b, whose
INITIATED marker makes a repeat a no-op. Phase A always clears stale state
first (ADR D4), so the guard that protects against a double prod deploy is the
INITIATED marker WRITTEN by Phase B — verify the auto path initiates exactly
once and a subsequent Phase B re-entry (duplicate confirm / reaper re-drive) is
a no-op."""
_label(monkeypatch, present=True)
initiate = MagicMock(return_value=(True, "ok"))
monkeypatch.setattr(stage_engine.self_deploy, "initiate_deploy", initiate)
tid = _make_task()
res = _advance(tid)
assert res.note == "self-deploy-initiated"
assert initiate.call_count == 1
assert self_deploy.has_marker("orchestrator", "ORCH-089", self_deploy.INITIATED)
# A repeat Phase B (e.g. duplicate Confirm Deploy webhook / reaper re-drive)
# with INITIATED already set is a no-op — no second prod deploy.
res2 = stage_engine._handle_self_deploy_phase_b(
tid, "orchestrator", "ORCH-089", "feature/ORCH-089-x",
stage_engine.AdvanceResult(from_stage="deploy"),
)
assert initiate.call_count == 1 # still exactly one
# --- TC-18: non-self / out of scope ----------------------------------------
def test_tc18_non_self_repo_no_phase_a(monkeypatch):
# For a non-self repo Phase A is not reached at all (self_deploy_applies False),
# so autoDeploy is a structural no-op. The edge advances normally to `deploy`.
_label(monkeypatch, present=True, applies=False)
initiate = MagicMock(return_value=(True, "ok"))
monkeypatch.setattr(stage_engine.self_deploy, "initiate_deploy", initiate)
tid = _make_task(repo="enduro-trails", wi="ET-1")
res = advance_stage(tid, "deploy-staging", "enduro-trails", "ET-1",
"feature/ORCH-089-x", finished_agent="deployer")
initiate.assert_not_called()
# No Phase A / Phase B for non-self repo.
assert res.note != "self-deploy-initiated"
# --- TC-19: transparency ---------------------------------------------------
def test_tc19_transparency_channels(monkeypatch, caplog):
_label(monkeypatch, present=True)
monkeypatch.setattr(stage_engine.self_deploy, "initiate_deploy",
MagicMock(return_value=(True, "ok")))
tid = _make_task()
import logging
with caplog.at_level(logging.INFO, logger="orchestrator.stage_engine"):
_advance(tid)
assert any("auto-confirmed" in r.message.lower() or "autoDeploy" in r.message
for r in caplog.records)
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-deploy Plane comment"