82 lines
2.7 KiB
Python
82 lines
2.7 KiB
Python
#!/usr/bin/env python3
|
|
"""Validate content catalogs against JSON Schema (CI + local).
|
|
|
|
Today: each object in content/skills/*.json top-level ``skills`` array vs
|
|
content/schemas/skill-def.schema.json. Extend this script as new catalogs land.
|
|
"""
|
|
|
|
from __future__ import annotations
|
|
|
|
import json
|
|
import sys
|
|
from pathlib import Path
|
|
|
|
from jsonschema import Draft202012Validator
|
|
|
|
REPO_ROOT = Path(__file__).resolve().parent.parent
|
|
SKILL_SCHEMA = REPO_ROOT / "content/schemas/skill-def.schema.json"
|
|
SKILLS_DIR = REPO_ROOT / "content/skills"
|
|
|
|
|
|
def main() -> int:
|
|
if not SKILL_SCHEMA.is_file():
|
|
print(f"error: missing schema {SKILL_SCHEMA}", file=sys.stderr)
|
|
return 1
|
|
|
|
schema = json.loads(SKILL_SCHEMA.read_text(encoding="utf-8"))
|
|
validator = Draft202012Validator(schema)
|
|
|
|
if not SKILLS_DIR.is_dir():
|
|
print(f"error: missing directory {SKILLS_DIR}", file=sys.stderr)
|
|
return 1
|
|
|
|
json_files = sorted(SKILLS_DIR.glob("*.json"))
|
|
if not json_files:
|
|
print(f"error: no JSON files under {SKILLS_DIR}", file=sys.stderr)
|
|
return 1
|
|
|
|
seen_ids: dict[str, str] = {}
|
|
errors = 0
|
|
|
|
for path in json_files:
|
|
data = json.loads(path.read_text(encoding="utf-8"))
|
|
skills = data.get("skills")
|
|
if not isinstance(skills, list):
|
|
print(f"error: {path.relative_to(REPO_ROOT)}: expected top-level 'skills' array", file=sys.stderr)
|
|
errors += 1
|
|
continue
|
|
for i, row in enumerate(skills):
|
|
if not isinstance(row, dict):
|
|
print(f"error: {path.relative_to(REPO_ROOT)}: skills[{i}] must be an object", file=sys.stderr)
|
|
errors += 1
|
|
continue
|
|
for err in sorted(validator.iter_errors(row), key=lambda e: e.path):
|
|
loc = ".".join(str(p) for p in err.path) or "(root)"
|
|
print(
|
|
f"error: {path.relative_to(REPO_ROOT)} skills[{i}] {loc}: {err.message}",
|
|
file=sys.stderr,
|
|
)
|
|
errors += 1
|
|
sid = row.get("id")
|
|
if isinstance(sid, str):
|
|
prev = seen_ids.get(sid)
|
|
if prev:
|
|
print(
|
|
f"error: duplicate skill id {sid!r} in {prev} and {path.relative_to(REPO_ROOT)}",
|
|
file=sys.stderr,
|
|
)
|
|
errors += 1
|
|
else:
|
|
seen_ids[sid] = str(path.relative_to(REPO_ROOT))
|
|
|
|
if errors:
|
|
print(f"content validation failed with {errors} error(s)", file=sys.stderr)
|
|
return 1
|
|
|
|
print(f"content OK: {len(json_files)} skill catalog file(s), {len(seen_ids)} unique skill id(s)")
|
|
return 0
|
|
|
|
|
|
if __name__ == "__main__":
|
|
raise SystemExit(main())
|