#!/usr/bin/env python3
"""Build site/data.json from data/reports/*.yml.

Every report is validated against data/schema.json before being included —
a malformed report fails the build loudly instead of shipping a broken row
to the site (same idea as archy's own scripts/validate-app-manifest.sh:
fail the build, don't ship the drift).
"""
import json
import sys
from pathlib import Path

import yaml
from jsonschema import Draft7Validator

ROOT = Path(__file__).resolve().parent.parent
REPORTS_DIR = ROOT / "data" / "reports"
SCHEMA_PATH = ROOT / "data" / "schema.json"
OUT_PATH = ROOT / "site" / "data.json"


def main() -> int:
    schema = json.loads(SCHEMA_PATH.read_text())
    validator = Draft7Validator(schema)

    reports = []
    errors = []

    for path in sorted(REPORTS_DIR.glob("*.yml")):
        try:
            data = yaml.safe_load(path.read_text())
        except yaml.YAMLError as e:
            errors.append(f"{path.name}: invalid YAML — {e}")
            continue

        report_errors = sorted(validator.iter_errors(data), key=lambda e: e.path)
        if report_errors:
            for err in report_errors:
                loc = ".".join(str(p) for p in err.path) or "(root)"
                errors.append(f"{path.name}: {loc}: {err.message}")
            continue

        data["_source_file"] = path.name
        reports.append(data)

    if errors:
        print(f"Build failed — {len(errors)} error(s):", file=sys.stderr)
        for e in errors:
            print(f"  {e}", file=sys.stderr)
        return 1

    reports.sort(key=lambda r: (r["device_model"], r["tested_date"]))
    OUT_PATH.write_text(json.dumps(reports, indent=2) + "\n")
    print(f"Wrote {len(reports)} report(s) to {OUT_PATH.relative_to(ROOT)}")
    return 0


if __name__ == "__main__":
    sys.exit(main())