neon-sprawl/scripts/validate_content.py

120 lines
4.4 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"
# Slice 1 prototype lock (NEO-33): exact ids + category coverage after schema passes.
PROTOTYPE_SLICE1_SKILL_IDS = frozenset({"salvage", "refine", "intrusion"})
def _prototype_slice1_gate(seen_ids: dict[str, str], id_to_category: dict[str, str]) -> str | None:
"""Return a human-readable error if Slice 1 contract fails, else None."""
ids = frozenset(seen_ids.keys())
if ids != PROTOTYPE_SLICE1_SKILL_IDS:
return (
"error: prototype Slice 1 expects exactly skill ids "
f"{sorted(PROTOTYPE_SLICE1_SKILL_IDS)!r}, got {sorted(ids)!r}"
)
cats = set(id_to_category.values())
if "gather" not in cats or "tech" not in cats:
return (
"error: prototype Slice 1 requires at least one gather and one tech skill; "
f"categories seen: {sorted(cats)!r}"
)
if "process" not in cats and "make" not in cats:
return (
"error: prototype Slice 1 requires at least one process or make skill; "
f"categories seen: {sorted(cats)!r}"
)
return None
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] = {}
id_to_category: 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
row_schema_errors = 0
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,
)
row_schema_errors += 1
errors += 1
sid = row.get("id")
# Reserve seen_ids (and duplicate detection) only for schema-clean rows so
# fixing a broken row does not surface a spurious duplicate vs an earlier invalid row.
if isinstance(sid, str) and row_schema_errors == 0:
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))
cat = row.get("category")
if isinstance(cat, str):
id_to_category[sid] = cat
if errors:
print(f"content validation failed with {errors} error(s)", file=sys.stderr)
return 1
slice1_err = _prototype_slice1_gate(seen_ids, id_to_category)
if slice1_err:
print(slice1_err, 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())