pegasus/scripts/publish_test_metrics.py

314 lines
11 KiB
Python
Executable File

#!/usr/bin/env python3
"""Publish Pegasus test-suite results to Prometheus via Pushgateway.
Inputs:
- Backend JUnit XML and frontend JUnit XML
- Backend/frontend coverage summaries
Outputs pushed:
- platform_quality_gate_runs_total{suite="pegasus",status="ok|failed"}
- pegasus_quality_gate_tests_total{suite="pegasus",result=*}
- pegasus_quality_gate_coverage_percent{suite="pegasus"}
"""
from __future__ import annotations
import json
import os
import urllib.request
import xml.etree.ElementTree as ET
from pathlib import Path
SOURCE_SUFFIXES = {".go", ".py", ".js", ".mjs", ".ts", ".tsx", ".json", ".yaml", ".yml", ".sh"}
SKIP_DIRS = {".git", ".venv", "venv", "node_modules", "build", "dist", "__pycache__", ".pytest_cache", "frontend-coverage"}
def _escape_label(value: str) -> str:
return value.replace("\\", "\\\\").replace("\n", "\\n").replace('"', '\\"')
def _label_str(labels: dict[str, str]) -> str:
parts = [f'{key}="{_escape_label(val)}"' for key, val in labels.items() if val]
return "{" + ",".join(parts) + "}" if parts else ""
def _read_text(path: Path) -> str:
if not path.exists():
return ""
return path.read_text(encoding="utf-8")
def _as_int(node: ET.Element, name: str) -> int:
raw = node.attrib.get(name) or "0"
try:
return int(float(raw))
except ValueError:
return 0
def _load_junit(path: Path) -> dict[str, int]:
if not path.exists():
return {"tests": 0, "failures": 0, "errors": 0, "skipped": 0}
tree = ET.parse(path)
root = tree.getroot()
suites: list[ET.Element]
if root.tag == "testsuite":
suites = [root]
elif root.tag == "testsuites":
suites = list(root.findall("testsuite"))
else:
suites = []
totals = {"tests": 0, "failures": 0, "errors": 0, "skipped": 0}
for suite in suites:
totals["tests"] += _as_int(suite, "tests")
totals["failures"] += _as_int(suite, "failures")
totals["errors"] += _as_int(suite, "errors")
totals["skipped"] += _as_int(suite, "skipped")
return totals
def _load_backend_coverage_percent(path: Path) -> float:
if not path.exists():
return 0.0
try:
return float(path.read_text(encoding="utf-8").strip())
except ValueError:
return 0.0
def _load_frontend_coverage_percent(path: Path) -> float:
if not path.exists():
return 0.0
payload = json.loads(path.read_text(encoding="utf-8"))
total = payload.get("total") or {}
lines = total.get("lines") or {}
pct = lines.get("pct")
if isinstance(pct, (int, float)):
return float(pct)
return 0.0
def _read_test_exit_code(path: Path) -> int:
if not path.exists():
return 1
raw = path.read_text(encoding="utf-8").strip()
try:
return int(raw)
except ValueError:
return 1
def _load_gate_summary(path: Path) -> dict[str, object]:
if not path.exists():
return {"ok": False, "issues": []}
try:
return json.loads(path.read_text(encoding="utf-8"))
except Exception:
return {"ok": False, "issues": []}
def _count_lines_over_limit(root: Path, *, max_lines: int = 500) -> int:
count = 0
for path in root.rglob("*"):
if not path.is_file():
continue
if any(part in SKIP_DIRS for part in path.parts):
continue
if path.name != "Jenkinsfile" and path.suffix.lower() not in SOURCE_SUFFIXES:
continue
try:
with path.open("r", encoding="utf-8", errors="ignore") as handle:
lines = sum(1 for _ in handle)
except OSError:
continue
if lines > max_lines:
count += 1
return count
def _fetch_existing_counter(pushgateway_url: str, metric: str, labels: dict[str, str]) -> float:
text = _read_http(f"{pushgateway_url.rstrip('/')}/metrics")
if not text:
return 0.0
for line in text.splitlines():
if not line.startswith(metric + "{"):
continue
if any(f'{k}="{v}"' not in line for k, v in labels.items()):
continue
parts = line.split()
if len(parts) < 2:
continue
try:
return float(parts[1])
except ValueError:
return 0.0
return 0.0
def _read_http(url: str) -> str:
try:
with urllib.request.urlopen(url, timeout=10) as resp:
return resp.read().decode("utf-8", errors="replace")
except Exception:
return ""
def _post_text(url: str, payload: str) -> None:
req = urllib.request.Request(
url,
data=payload.encode("utf-8"),
method="POST",
headers={"Content-Type": "text/plain"},
)
with urllib.request.urlopen(req, timeout=10) as resp:
if resp.status >= 400:
raise RuntimeError(f"push failed status={resp.status}")
def main() -> int:
suite = os.getenv("SUITE_NAME", "pegasus")
pushgateway_url = os.getenv(
"PUSHGATEWAY_URL", "http://platform-quality-gateway.monitoring.svc.cluster.local:9091"
)
backend_junit = Path(os.getenv("BACKEND_JUNIT_XML", "build/junit-backend.xml"))
frontend_junit = Path(os.getenv("FRONTEND_JUNIT_XML", "build/junit-frontend.xml"))
backend_cov = Path(os.getenv("BACKEND_COVERAGE_PERCENT_FILE", "build/coverage-backend-percent.txt"))
frontend_cov = Path(
os.getenv("FRONTEND_COVERAGE_JSON", "build/frontend-coverage/coverage-summary.json")
)
backend_rc_file = Path(os.getenv("BACKEND_TEST_RC_FILE", "build/backend-test.rc"))
frontend_rc_file = Path(os.getenv("FRONTEND_TEST_RC_FILE", "build/frontend-test.rc"))
gate_summary = _load_gate_summary(Path(os.getenv("GATE_SUMMARY_FILE", "build/gate-summary.json")))
repo_root = Path(__file__).resolve().parents[1]
b = _load_junit(backend_junit)
f = _load_junit(frontend_junit)
totals = {
"tests": b["tests"] + f["tests"],
"failures": b["failures"] + f["failures"],
"errors": b["errors"] + f["errors"],
"skipped": b["skipped"] + f["skipped"],
}
passed = max(totals["tests"] - totals["failures"] - totals["errors"] - totals["skipped"], 0)
backend_pct = _load_backend_coverage_percent(backend_cov)
frontend_pct = _load_frontend_coverage_percent(frontend_cov)
coverage_pct = (backend_pct + frontend_pct) / 2 if (backend_pct or frontend_pct) else 0.0
over_500 = _count_lines_over_limit(repo_root)
backend_rc = _read_test_exit_code(backend_rc_file)
frontend_rc = _read_test_exit_code(frontend_rc_file)
backend_suite_result = "passed" if backend_rc == 0 else "failed"
frontend_suite_result = "passed" if frontend_rc == 0 else "failed"
branch = os.getenv("BRANCH_NAME", "")
build_number = os.getenv("BUILD_NUMBER", "")
commit = os.getenv("GIT_COMMIT", "")
labels = {
"suite": suite,
"branch": branch,
"build_number": build_number,
"commit": commit,
}
gate_ok = bool(gate_summary.get("ok"))
gate_issues = gate_summary.get("issues") or []
outcome = (
"ok"
if gate_ok
and backend_rc == 0
and frontend_rc == 0
and totals["tests"] > 0
and totals["failures"] == 0
and totals["errors"] == 0
else "failed"
)
job_name = "platform-quality-ci"
ok_count = _fetch_existing_counter(
pushgateway_url,
"platform_quality_gate_runs_total",
{"job": job_name, "suite": suite, "status": "ok"},
)
failed_count = _fetch_existing_counter(
pushgateway_url,
"platform_quality_gate_runs_total",
{"job": job_name, "suite": suite, "status": "failed"},
)
if outcome == "ok":
ok_count += 1
else:
failed_count += 1
tests_check = "ok" if outcome == "ok" else "failed"
coverage_check = "ok" if coverage_pct >= 95.0 else "failed"
loc_check = "ok" if over_500 == 0 else "failed"
gate_check = "ok" if gate_ok else "failed"
payload_lines = [
"# TYPE platform_quality_gate_runs_total counter",
f'platform_quality_gate_runs_total{{suite="{suite}",status="ok"}} {ok_count:.0f}',
f'platform_quality_gate_runs_total{{suite="{suite}",status="failed"}} {failed_count:.0f}',
"# TYPE pegasus_test_suite_result gauge",
f'pegasus_test_suite_result{{suite="backend",status="{backend_suite_result}"}} 1',
f'pegasus_test_suite_result{{suite="frontend",status="{frontend_suite_result}"}} 1',
"# TYPE pegasus_quality_gate_tests_total gauge",
f'pegasus_quality_gate_tests_total{{suite="{suite}",result="passed"}} {passed}',
f'pegasus_quality_gate_tests_total{{suite="{suite}",result="failed"}} {totals["failures"]}',
f'pegasus_quality_gate_tests_total{{suite="{suite}",result="error"}} {totals["errors"]}',
f'pegasus_quality_gate_tests_total{{suite="{suite}",result="skipped"}} {totals["skipped"]}',
"# TYPE pegasus_quality_gate_coverage_percent gauge",
f'pegasus_quality_gate_coverage_percent{{suite="{suite}"}} {coverage_pct:.3f}',
"# TYPE pegasus_quality_gate_status gauge",
f'pegasus_quality_gate_status{{suite="{suite}",result="{"ok" if gate_ok else "failed"}"}} 1',
"# TYPE pegasus_quality_gate_issues_total gauge",
f'pegasus_quality_gate_issues_total{{suite="{suite}"}} {len(gate_issues)}',
"# TYPE platform_quality_gate_workspace_line_coverage_percent gauge",
f'platform_quality_gate_workspace_line_coverage_percent{{suite="{suite}"}} {coverage_pct:.3f}',
"# TYPE platform_quality_gate_source_lines_over_500_total gauge",
f'platform_quality_gate_source_lines_over_500_total{{suite="{suite}"}} {over_500}',
"# TYPE pegasus_quality_gate_checks_total gauge",
f'pegasus_quality_gate_checks_total{{suite="{suite}",check="tests",result="{tests_check}"}} 1',
f'pegasus_quality_gate_checks_total{{suite="{suite}",check="coverage",result="{coverage_check}"}} 1',
f'pegasus_quality_gate_checks_total{{suite="{suite}",check="loc",result="{loc_check}"}} 1',
f'pegasus_quality_gate_checks_total{{suite="{suite}",check="gate",result="{gate_check}"}} 1',
"# TYPE pegasus_quality_gate_build_info gauge",
f"pegasus_quality_gate_build_info{_label_str(labels)} 1",
]
payload = "\n".join(payload_lines) + "\n"
push_url = f"{pushgateway_url.rstrip('/')}/metrics/job/{job_name}/suite/{suite}"
_post_text(push_url, payload)
summary = {
"suite": suite,
"tests_total": totals["tests"],
"tests_passed": passed,
"tests_failed": totals["failures"],
"tests_errors": totals["errors"],
"tests_skipped": totals["skipped"],
"coverage_percent": round(coverage_pct, 3),
"source_lines_over_500": over_500,
"outcome": outcome,
"backend_rc": backend_rc,
"frontend_rc": frontend_rc,
"ok_counter": ok_count,
"failed_counter": failed_count,
}
Path("build/metrics-summary.json").write_text(json.dumps(summary, indent=2), encoding="utf-8")
print(json.dumps(summary, indent=2))
return 0
if __name__ == "__main__":
try:
raise SystemExit(main())
except Exception as exc:
print(f"metrics push failed: {exc}")
raise