pryscraper/routers/quality.py
cryptorugmunch 2f1eec2f78
Some checks failed
CI / lint (pull_request) Successful in 33s
CI / typecheck (pull_request) Failing after 1m42s
CI / test (pull_request) Failing after 2m33s
CI / security (pull_request) Failing after 39s
CI / gitleaks (pull_request) Successful in 36s
CI / commitlint (pull_request) Failing after 10s
style: ruff format
2026-07-03 18:12:36 +02:00

90 lines
2.6 KiB
Python

"""Pry — Quality router (remaining api.py routes).
Auto-extracted from api.py during the router-split refactor.
"""
# SPDX-License-Identifier: MIT
# Copyright (c) 2026 Rug Munch Media LLC
from __future__ import annotations
import json
import logging
import os
from typing import Any
from fastapi import APIRouter, Body
logger = logging.getLogger(__name__)
router = APIRouter(tags=["Quality"])
@router.post(
"/v1/quality/check", tags=["Quality"], summary="Run data quality check on extraction results"
)
async def quality_check(
url: str = Body(...),
data: dict[str, Any] = Body(...),
schema: dict[str, Any] | None = Body(None),
expected_types: dict[str, str] | None = Body(None),
) -> dict[str, Any]:
"""Run a full data quality check on extraction results.
Metrics:
- Completeness: what % of expected fields have values
- Schema adherence: field types match expectations
- Freshness: how old is the data
- Anomaly detection: what changed since last extraction
Use this to validate data BEFORE sending to downstream systems.
"""
from quality import run_quality_check
# Convert string type names to actual types
type_map: dict[str, type] = {
"str": str,
"int": int,
"float": float,
"bool": bool,
"list": list,
"dict": dict,
"None": type(None),
}
resolved_types: dict[str, type] | None = None
if expected_types:
resolved_types = {}
for field, type_name in expected_types.items():
resolved_types[field] = type_map.get(type_name, str)
result = await run_quality_check(
url=url,
data=data,
schema=schema,
expected_types=resolved_types,
)
return {"success": True, "data": result}
@router.get(
"/v1/quality/stats", tags=["Quality"], summary="Get quality statistics for all checked URLs"
)
async def quality_stats() -> dict[str, Any]:
"""Get aggregate quality statistics across all checked URLs."""
from quality import QUALITY_DIR
stats: list[dict[str, Any]] = []
for path in sorted(QUALITY_DIR.glob("*.json"), key=os.path.getmtime, reverse=True)[:50]:
try:
data = json.loads(path.read_text())
stats.append(
{
"url": data.get("url", ""),
"checked_at": data.get("checked_at", ""),
"size_bytes": path.stat().st_size,
}
)
except (json.JSONDecodeError, OSError):
continue
return {"success": True, "data": {"total_checked": len(stats), "recent": stats}}