Add governed source freshness audit
GeoIntel CI / docs-smoke (push) Canceled after 0s
GeoIntel CI / contract-smoke (push) Canceled after 0s

This commit is contained in:
Codex
2026-07-16 15:05:42 +02:00
parent 2723043e3d
commit 1e0e7b6499
22 changed files with 1148 additions and 0 deletions
+98
View File
@@ -0,0 +1,98 @@
#!/usr/bin/env python3
"""Read-only source freshness and local integrity audit for one GeoIntel project."""
from __future__ import annotations
import argparse
import json
import sys
from pathlib import Path
from urllib.error import HTTPError, URLError
from urllib.request import Request, urlopen
def parse_args() -> argparse.Namespace:
parser = argparse.ArgumentParser(
description="Inspect persisted source versions and storage evidence without refreshing source data."
)
parser.add_argument("--project-id", required=True, help="GeoIntel project UUID")
parser.add_argument("--api-url", default="http://127.0.0.1/api/v1", help="API base URL ending in /api/v1")
parser.add_argument("--timeout", type=float, default=30.0, help="HTTP timeout in seconds")
parser.add_argument("--json", action="store_true", help="Print the canonical report data as JSON")
parser.add_argument("--output", type=Path, help="Optional path for a JSON evidence copy")
parser.add_argument(
"--fail-on",
choices=("never", "integrity", "due", "attention"),
default="integrity",
help="Non-zero exit policy for cron or release automation",
)
return parser.parse_args()
def fetch_report(api_url: str, project_id: str, timeout: float) -> dict:
endpoint = f"{api_url.rstrip('/')}/projects/{project_id}/datasets/source-freshness"
request = Request(endpoint, headers={"Accept": "application/json", "User-Agent": "GeoIntel-source-audit/1.0"})
try:
with urlopen(request, timeout=timeout) as response:
payload = json.load(response)
except HTTPError as exc:
body = exc.read().decode("utf-8", errors="replace")
raise RuntimeError(f"GeoIntel API returned HTTP {exc.code}: {body}") from exc
except URLError as exc:
raise RuntimeError(f"GeoIntel API is unreachable: {exc.reason}") from exc
if not isinstance(payload, dict) or not isinstance(payload.get("data"), dict):
raise RuntimeError("Response is not a canonical GeoIntel data envelope")
return payload["data"]
def print_text(report: dict) -> None:
summary = report.get("summary", {})
print(
"GeoIntel source audit: "
f"{summary.get('source_count', 0)} sources, "
f"{summary.get('dataset_count', 0)} datasets, "
f"{summary.get('due_count', 0)} due, "
f"{summary.get('review_required_count', 0)} review, "
f"{summary.get('integrity_issue_count', 0)} integrity issues"
)
for item in report.get("items", []):
status = item.get("status", "unknown")
if status not in {"current", "local"} or item.get("integrity", {}).get("missing_version_count", 0):
print(f"- {status:15} {item.get('display_name', item.get('source_name'))}: {item.get('reason', '')}")
print("No external catalog was queried and no dataset was modified.")
def should_fail(report: dict, fail_on: str) -> bool:
summary = report.get("summary", {})
integrity = int(summary.get("integrity_issue_count", 0))
due = int(summary.get("due_count", 0))
review = int(summary.get("review_required_count", 0))
if fail_on == "never":
return False
if fail_on == "integrity":
return integrity > 0
if fail_on == "due":
return integrity > 0 or due > 0
return integrity > 0 or due > 0 or review > 0
def main() -> int:
args = parse_args()
try:
report = fetch_report(args.api_url, args.project_id, args.timeout)
except RuntimeError as exc:
print(str(exc), file=sys.stderr)
return 2
serialized = json.dumps(report, indent=2, sort_keys=True)
if args.output:
args.output.parent.mkdir(parents=True, exist_ok=True)
args.output.write_text(serialized + "\n", encoding="utf-8")
if args.json:
print(serialized)
else:
print_text(report)
return 1 if should_fail(report, args.fail_on) else 0
if __name__ == "__main__":
raise SystemExit(main())