fix: resolve conflicts for PR #358 against latest master
Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>
This commit is contained in:
@@ -1242,6 +1242,102 @@ def assess_workspace_mutation_consistency(report_text, observed_commands=None):
|
||||
}
|
||||
|
||||
|
||||
# ── Already-landed report wording (Issue #298) ───────────────────────────────
|
||||
#
|
||||
# An already-landed PR is reconciliation-only: it must never be called
|
||||
# eligible, its head SHA must never be called reviewed, and the legacy
|
||||
# eligible/reviewed/scratch-worktree handoff fields must not appear.
|
||||
# Reviewed-head claims in any report require validation + diff review to
|
||||
# have actually passed.
|
||||
|
||||
ALREADY_LANDED_FORBIDDEN_PHRASES = (
|
||||
"oldest eligible pr",
|
||||
"next eligible pr",
|
||||
"pinned reviewed head",
|
||||
"scratch worktree used",
|
||||
)
|
||||
|
||||
ALREADY_LANDED_ELIGIBILITY_CLASS = "ALREADY_LANDED_RECONCILE_REQUIRED"
|
||||
|
||||
|
||||
def assess_already_landed_report_wording(report_text, *, already_landed,
|
||||
review_validated=False):
|
||||
"""#298: reject eligible/reviewed wording for already-landed PRs.
|
||||
|
||||
*already_landed* states whether the already-landed gate fired for the
|
||||
selected PR. *review_validated* states whether validation and diff
|
||||
review actually passed this session; only then may a reviewed head
|
||||
SHA be populated in a non-already-landed report.
|
||||
|
||||
Returns {'complete', 'downgraded', 'reasons'}; fails closed.
|
||||
"""
|
||||
text_lower = (report_text or "").lower()
|
||||
fields = _report_labeled_fields(report_text)
|
||||
reasons = []
|
||||
|
||||
reviewed_head = fields.get("reviewed head sha")
|
||||
reviewed_head_populated = bool(
|
||||
reviewed_head and not reviewed_head.strip().lower().startswith("none")
|
||||
)
|
||||
pinned_head_present = "pinned reviewed head" in fields
|
||||
|
||||
if already_landed:
|
||||
for phrase in ALREADY_LANDED_FORBIDDEN_PHRASES:
|
||||
if phrase in text_lower:
|
||||
reasons.append(
|
||||
f"already-landed report must not use '{phrase}'; the PR "
|
||||
"is reconciliation-only, not review/merge eligible"
|
||||
)
|
||||
|
||||
eligibility = fields.get("eligibility class", "")
|
||||
if ALREADY_LANDED_ELIGIBILITY_CLASS not in eligibility.upper():
|
||||
reasons.append(
|
||||
"already-landed report missing 'Eligibility class: "
|
||||
f"{ALREADY_LANDED_ELIGIBILITY_CLASS}'"
|
||||
)
|
||||
|
||||
for required in ("oldest open pr requiring action",
|
||||
"candidate head sha"):
|
||||
if required not in fields:
|
||||
reasons.append(
|
||||
f"already-landed report missing '{required}' field"
|
||||
)
|
||||
|
||||
if "reviewed head sha" not in fields:
|
||||
reasons.append(
|
||||
"already-landed report must state 'Reviewed head SHA: none'"
|
||||
)
|
||||
elif reviewed_head_populated:
|
||||
reasons.append(
|
||||
"already-landed report must not claim a reviewed head SHA; "
|
||||
"no validation or diff review passed for this PR"
|
||||
)
|
||||
|
||||
worktree_used = fields.get("review worktree used", "")
|
||||
if worktree_used.strip().lower() not in ("false", "no"):
|
||||
reasons.append(
|
||||
"already-landed report must state 'Review worktree used: "
|
||||
"false'"
|
||||
)
|
||||
elif not review_validated:
|
||||
if pinned_head_present:
|
||||
reasons.append(
|
||||
"'Pinned reviewed head' populated before validation and "
|
||||
"diff review passed"
|
||||
)
|
||||
if reviewed_head_populated:
|
||||
reasons.append(
|
||||
"reviewed head SHA populated before validation and diff "
|
||||
"review passed"
|
||||
)
|
||||
|
||||
return {
|
||||
"complete": not reasons,
|
||||
"downgraded": bool(reasons),
|
||||
"reasons": reasons,
|
||||
}
|
||||
|
||||
|
||||
def assess_role_boundary(proof=None, *, task_role=None, namespaces_used=None,
|
||||
justification=None):
|
||||
"""Assess reviewer/author role separation for blind queue workflows.
|
||||
@@ -1422,6 +1518,7 @@ def build_final_report(checkout_proof, inventory, validation, contamination,
|
||||
report_text=None, review_decision_lock=None,
|
||||
controller_handoff=None, capability_proof=None,
|
||||
sweep_proof=None, worktree_proof=None,
|
||||
validation_environment=None,
|
||||
queue_ordering=None,
|
||||
baseline_validation=None, project_root=None):
|
||||
"""Required behavior 6 + acceptance criteria: one report, distinct proofs.
|
||||
@@ -1450,6 +1547,22 @@ def build_final_report(checkout_proof, inventory, validation, contamination,
|
||||
)
|
||||
|
||||
empty_queue_report = assess_empty_queue_report(report_text)
|
||||
if validation_environment is None and report_text:
|
||||
validation_env_proof = assess_validation_environment_proof(
|
||||
report_text=report_text,
|
||||
)
|
||||
elif validation_environment is not None:
|
||||
validation_env_proof = assess_validation_environment_proof(
|
||||
report_text=report_text,
|
||||
validation_environment=validation_environment,
|
||||
)
|
||||
else:
|
||||
validation_env_proof = {
|
||||
"proven": True,
|
||||
"block": False,
|
||||
"reasons": [],
|
||||
"violations": [],
|
||||
}
|
||||
if queue_ordering is None and report_text:
|
||||
queue_ordering_proof = assess_queue_ordering_proof(
|
||||
report_text=report_text,
|
||||
@@ -1617,6 +1730,11 @@ def build_final_report(checkout_proof, inventory, validation, contamination,
|
||||
"empty-queue report missing or failed trust-gate proof (#198)"
|
||||
)
|
||||
downgrade_reasons.extend(empty_queue_report.get("reasons", []))
|
||||
if not validation_env_proof.get("proven"):
|
||||
downgrade_reasons.append(
|
||||
"validation environment proof missing or failed (#311)"
|
||||
)
|
||||
downgrade_reasons.extend(validation_env_proof.get("reasons", []))
|
||||
if not queue_ordering_proof.get("proven"):
|
||||
downgrade_reasons.append(
|
||||
"queue ordering proof missing or failed (#321)"
|
||||
@@ -1648,11 +1766,13 @@ def build_final_report(checkout_proof, inventory, validation, contamination,
|
||||
# #179: no merge without a proven final live-state recheck.
|
||||
and live_state_proven
|
||||
and worktree_proven
|
||||
and validation_env_proof.get("proven")
|
||||
and queue_ordering_proof.get("proven")
|
||||
and baseline_validation.get("proven")
|
||||
)
|
||||
|
||||
violations = []
|
||||
violations.extend(validation_env_proof.get("violations", []))
|
||||
violations.extend(queue_ordering_proof.get("violations", []))
|
||||
violations.extend(baseline_validation.get("violations", []))
|
||||
if merge_performed and not merge_allowed:
|
||||
@@ -1706,6 +1826,12 @@ def build_final_report(checkout_proof, inventory, validation, contamination,
|
||||
else True
|
||||
),
|
||||
"empty_queue_trust_gate_status": empty_queue_report.get("status"),
|
||||
"validation_environment_proven": bool(
|
||||
validation_env_proof.get("proven")
|
||||
),
|
||||
"validation_environment_violations": list(
|
||||
validation_env_proof.get("violations") or []
|
||||
),
|
||||
"queue_ordering_proven": bool(queue_ordering_proof.get("proven")),
|
||||
"queue_ordering_policy": queue_ordering_proof.get("policy"),
|
||||
"queue_ordering_violations": list(
|
||||
@@ -3322,6 +3448,192 @@ def build_review_mutation_proof(run_log: list[dict]) -> dict:
|
||||
}
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Validation environment proof (#311)
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
_VALIDATION_CLAIM_RE = re.compile(
|
||||
r"(?:validation\s+command|pytest\s+(?:executed|passed|failed)|"
|
||||
r"\d+\s+passed|\d+\s+failed|tests?\s+passed)",
|
||||
re.IGNORECASE,
|
||||
)
|
||||
|
||||
_VALIDATION_CMD_LINE_RE = re.compile(
|
||||
r"validation\s+command\s*:\s*(.+)",
|
||||
re.IGNORECASE,
|
||||
)
|
||||
|
||||
_VALIDATION_CWD_LINE_RE = re.compile(
|
||||
r"working\s+directory\s*:\s*(\S+)",
|
||||
re.IGNORECASE,
|
||||
)
|
||||
|
||||
_WHICH_PYTEST_LINE_RE = re.compile(
|
||||
r"which\s+pytest\s*:\s*(.+)",
|
||||
re.IGNORECASE,
|
||||
)
|
||||
|
||||
_PYTEST_VERSION_LINE_RE = re.compile(
|
||||
r"pytest\s+--version\s*:\s*(.+)",
|
||||
re.IGNORECASE,
|
||||
)
|
||||
|
||||
_BARE_PYTEST_CMD_RE = re.compile(
|
||||
r"^(?:python\s+-m\s+)?pytest(?:\s|$)",
|
||||
re.IGNORECASE,
|
||||
)
|
||||
|
||||
_VAGUE_PYTEST_CLAIM_RE = re.compile(
|
||||
r"pytest\s+(?:executed|passed|failed)\b",
|
||||
re.IGNORECASE,
|
||||
)
|
||||
|
||||
|
||||
def _parse_validation_command(report_text: str) -> str:
|
||||
for line in (report_text or "").splitlines():
|
||||
match = _VALIDATION_CMD_LINE_RE.search(line)
|
||||
if match:
|
||||
return match.group(1).strip().rstrip(".")
|
||||
return ""
|
||||
|
||||
|
||||
def _parse_validation_cwd(report_text: str) -> str:
|
||||
for line in (report_text or "").splitlines():
|
||||
match = _VALIDATION_CWD_LINE_RE.search(line)
|
||||
if match:
|
||||
return match.group(1).strip().rstrip(",.;")
|
||||
return ""
|
||||
|
||||
|
||||
def _is_bare_pytest_command(command: str) -> bool:
|
||||
text = (command or "").strip()
|
||||
if not text:
|
||||
return False
|
||||
if "/" in text or "\\" in text:
|
||||
return False
|
||||
return bool(_BARE_PYTEST_CMD_RE.match(text))
|
||||
|
||||
|
||||
def assess_validation_environment_proof(
|
||||
*,
|
||||
report_text: str | None = None,
|
||||
validation_environment: dict | None = None,
|
||||
) -> dict:
|
||||
"""#311: reviewer reports must include exact validation command/environment."""
|
||||
text = report_text or ""
|
||||
lower = text.lower()
|
||||
env = validation_environment or {}
|
||||
reasons: list[str] = []
|
||||
violations: list[str] = []
|
||||
|
||||
claims_validation = bool(_VALIDATION_CLAIM_RE.search(text))
|
||||
if not claims_validation and not env.get("command"):
|
||||
return {
|
||||
"proven": True,
|
||||
"block": False,
|
||||
"claims_validation": False,
|
||||
"reasons": [],
|
||||
"violations": [],
|
||||
"safe_next_action": "proceed",
|
||||
}
|
||||
|
||||
command = (env.get("command") or _parse_validation_command(text)).strip()
|
||||
working_directory = (
|
||||
env.get("working_directory") or env.get("cwd")
|
||||
or _parse_validation_cwd(text)
|
||||
).strip()
|
||||
|
||||
if not command:
|
||||
reasons.append(
|
||||
"validation result claimed without exact validation command (#311)"
|
||||
)
|
||||
elif _VAGUE_PYTEST_CLAIM_RE.search(text) and command == "":
|
||||
violations.append(
|
||||
"vague pytest summary without exact validation command (#311)"
|
||||
)
|
||||
|
||||
if not working_directory:
|
||||
reasons.append(
|
||||
"validation result claimed without working directory (#311)"
|
||||
)
|
||||
|
||||
has_result = any(
|
||||
token in lower
|
||||
for token in ("passed", "failed", "skipped", "error")
|
||||
) or any(
|
||||
env.get(key) is not None
|
||||
for key in ("passed", "failed", "skipped", "result")
|
||||
)
|
||||
if not has_result:
|
||||
reasons.append(
|
||||
"validation result claimed without pass/fail result evidence (#311)"
|
||||
)
|
||||
|
||||
if command and _is_bare_pytest_command(command):
|
||||
which_pytest = (
|
||||
(env.get("which_pytest") or env.get("executable_path") or "")
|
||||
.strip()
|
||||
)
|
||||
if not which_pytest:
|
||||
for line in text.splitlines():
|
||||
match = _WHICH_PYTEST_LINE_RE.search(line)
|
||||
if match:
|
||||
which_pytest = match.group(1).strip()
|
||||
break
|
||||
if not which_pytest:
|
||||
reasons.append(
|
||||
"bare pytest command requires which-pytest/executable proof (#311)"
|
||||
)
|
||||
|
||||
pytest_version = (env.get("pytest_version") or "").strip()
|
||||
if not pytest_version:
|
||||
for line in text.splitlines():
|
||||
match = _PYTEST_VERSION_LINE_RE.search(line)
|
||||
if match:
|
||||
pytest_version = match.group(1).strip()
|
||||
break
|
||||
if not pytest_version:
|
||||
reasons.append(
|
||||
"bare pytest command requires pytest --version proof (#311)"
|
||||
)
|
||||
|
||||
venv_resolved = env.get("venv_resolved")
|
||||
if venv_resolved is None:
|
||||
venv_resolved = any(
|
||||
marker in lower
|
||||
for marker in (
|
||||
"project venv",
|
||||
"venv/bin/pytest",
|
||||
"resolves to the project venv",
|
||||
"venv resolved: yes",
|
||||
)
|
||||
)
|
||||
if not venv_resolved and which_pytest and "venv" not in which_pytest:
|
||||
reasons.append(
|
||||
"bare pytest command requires proof whether executable "
|
||||
"resolves to the project venv (#311)"
|
||||
)
|
||||
|
||||
proven = not reasons and not violations
|
||||
return {
|
||||
"proven": proven,
|
||||
"block": bool(violations),
|
||||
"claims_validation": True,
|
||||
"command": command or None,
|
||||
"working_directory": working_directory or None,
|
||||
"reasons": reasons,
|
||||
"violations": violations,
|
||||
"safe_next_action": (
|
||||
"report exact validation command, working directory, result, and "
|
||||
"for bare pytest also which-pytest, pytest --version, and venv proof"
|
||||
if not proven
|
||||
else "proceed"
|
||||
),
|
||||
}
|
||||
|
||||
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Reviewer queue ordering proof (#321)
|
||||
# ---------------------------------------------------------------------------
|
||||
@@ -4237,6 +4549,13 @@ def assess_git_ref_mutation_report(report_text, *, command_log=None):
|
||||
}
|
||||
|
||||
|
||||
def assess_merge_simulation_report(report_text, **kwargs):
|
||||
"""#317: classify local merge simulation as worktree/index mutations."""
|
||||
from reviewer_merge_simulation import assess_merge_simulation_report as _assess
|
||||
|
||||
return _assess(report_text, **kwargs)
|
||||
|
||||
|
||||
def assess_validation_integrity_report(report_text, **kwargs):
|
||||
"""#316: separate official PR-head validation from diagnostic experiments."""
|
||||
from reviewer_validation_integrity import assess_validation_integrity_report as _assess
|
||||
@@ -4258,6 +4577,15 @@ def assess_non_mergeable_skip_proof(report_text, **kwargs):
|
||||
return _assess(report_text, **kwargs)
|
||||
|
||||
|
||||
def assess_validation_worktree_edit_report(report_text, **kwargs):
|
||||
"""#315: block edits in PR validation worktrees; require diagnostic separation."""
|
||||
from reviewer_validation_worktree_edits import (
|
||||
assess_validation_worktree_edit_report as _assess,
|
||||
)
|
||||
|
||||
return _assess(report_text, **kwargs)
|
||||
|
||||
|
||||
def assess_worktree_ownership_report(report_text, **kwargs):
|
||||
"""#312: prove session-owned or safe-reuse worktree before reset/validation."""
|
||||
from reviewer_worktree_ownership import assess_worktree_ownership_report as _assess
|
||||
|
||||
Reference in New Issue
Block a user