Files
cdsl-cad/backend/engine/cdsl_engine/semantic_validation.py
T

295 lines
17 KiB
Python

"""Validation for the complete CDSL v1.1 semantic contract.
The current runtime accepts only a subset of this contract. Keeping this
validator separate lets import tooling preserve a SolidWorks feature history
without claiming that every feature can already be rebuilt locally.
"""
from __future__ import annotations
import json
import re
from functools import lru_cache
from pathlib import Path
from typing import Any
from jsonschema import Draft202012Validator
from .operation_contracts import materialized_feature_contracts
_ID = re.compile(r"^[A-Za-z0-9_-]{1,80}$")
def _mappings(value: Any):
"""Yield nested mapping values without treating selector-like data as text."""
if isinstance(value, dict):
yield value
for child in value.values():
yield from _mappings(child)
elif isinstance(value, list):
for child in value:
yield from _mappings(child)
@lru_cache(maxsize=1)
def _schema() -> dict[str, Any]:
path = Path(__file__).with_name("cdsl_schema.json")
schema = json.loads(path.read_text(encoding="utf-8"))
Draft202012Validator.check_schema(schema)
return schema
@lru_cache(maxsize=1)
def _validator() -> Draft202012Validator:
return Draft202012Validator(_schema())
@lru_cache(maxsize=1)
def _operation_contracts() -> dict[str, dict[str, Any]]:
path = Path(__file__).with_name("profile_schema.json")
return materialized_feature_contracts(json.loads(path.read_text(encoding="utf-8")))
def _schema_error(document: dict[str, Any]) -> str | None:
validator = _validator()
errors = sorted(validator.iter_errors(document), key=lambda error: (list(error.absolute_path), error.message))
if not errors:
return None
error = errors[0]
location = "$" + "".join(f"[{item}]" if isinstance(item, int) else f".{item}" for item in error.absolute_path)
return f"CDSL schema violation at {location}: {error.message}"
def validate_semantic_cdsl(cdsl: dict[str, Any]) -> dict[str, Any]:
"""Validate a CDSL document without invoking the rebuild compiler.
The return value is intentionally serializable so the batch converter can
write it unchanged into a per-model diagnostic file.
"""
if not isinstance(cdsl, dict):
raise ValueError("CDSL must be a JSON object")
if cdsl.get("schema") != "cad.cdsl.llm.v1":
raise ValueError("Unsupported CDSL schema")
schema_error = _schema_error(cdsl)
if schema_error:
raise ValueError(schema_error)
version = str(cdsl.get("schema_version") or "1.0.0")
if not re.fullmatch(r"1\.[0-9]+\.[0-9]+", version):
raise ValueError("schema_version must be a 1.x.y version")
version_numbers = tuple(int(component) for component in version.split("."))
if version_numbers >= (1, 1, 0) and cdsl.get("meta", {}).get("unit") != "mm":
raise ValueError("CDSL v1.1 requires meta.unit = 'mm'")
sketches = (cdsl.get("geometry") or {}).get("sketches") or []
sketch_ids = {str(sketch.get("id") or "") for sketch in sketches}
if len(sketch_ids) != len(sketches) or not all(_ID.fullmatch(item) for item in sketch_ids):
raise ValueError("Sketch ids must be unique valid CDSL identifiers")
for sketch in sketches:
profile = sketch.get("profile") or {}
if profile.get("type") != "planar_imprint":
continue
entities = profile.get("source_entities") or []
entity_ids = [str(entity.get("id") or "") for entity in entities if isinstance(entity, dict)]
if len(entity_ids) != len(entities) or len(set(entity_ids)) != len(entity_ids) or not all(entity_ids):
raise ValueError(f"Sketch {sketch.get('id')} planar_imprint source entity ids must be unique")
known = set(entity_ids)
for index, selection in enumerate(profile.get("selections") or []):
source = str(selection.get("source_entity_id") or "")
fragment = selection.get("fragment") or {}
anchor = str(fragment.get("anchor_entity_id") or "") if fragment else ""
if source not in known:
raise ValueError(f"Sketch {sketch.get('id')} planar_imprint selection {index} has an unknown source entity")
if fragment and (anchor not in known or anchor == source):
raise ValueError(f"Sketch {sketch.get('id')} planar_imprint selection {index} has an invalid fragment anchor")
feature_ids: set[str] = set()
preceding_features: dict[str, dict[str, Any]] = {}
deferred: list[str] = []
unresolved: list[dict[str, Any]] = []
contracts = _operation_contracts()
for feature in cdsl.get("features") or []:
fid = str(feature.get("id") or "")
if not _ID.fullmatch(fid) or fid in feature_ids:
raise ValueError("Feature ids must be unique valid CDSL identifiers")
for dependency in feature.get("depends_on") or []:
if dependency not in feature_ids:
raise ValueError(f"Feature {fid} has a forward or missing dependency: {dependency}")
sketch_id = feature.get("sketch_id")
if sketch_id is not None and str(sketch_id) not in sketch_ids:
raise ValueError(f"Feature {fid} refers to a missing sketch: {sketch_id}")
if version_numbers >= (1, 1, 0) and feature.get("execution_status") not in {"supported", "deferred"}:
raise ValueError(f"Feature {fid} must declare execution_status")
if feature.get("execution_status") == "deferred":
deferred.append(fid)
contract = contracts.get(str(feature.get("atomic_id") or "")) or {}
feature_selectors = feature.get("selectors") or []
output_role_selector_ids = {id(selector) for selector in feature_selectors if isinstance(selector, dict)}
for index, selector in enumerate(feature_selectors):
owner = selector.get("owner_feature_id")
binding_owner = selector.get("binding_feature_id")
if owner is not None and owner not in feature_ids and binding_owner not in feature_ids:
raise ValueError(f"Feature {fid} selector {index} has a forward or missing owner_feature_id")
if selector.get("output_role") is not None:
if contract.get("selector_slot") != "feature.selectors" or contract.get("selector_token_kind") != "face":
raise ValueError(f"Feature {fid} selector {index} cannot consume a feature output role")
if selector.get("kind") != "face" or not owner or owner not in feature_ids:
raise ValueError(f"Feature {fid} selector {index} output role requires a preceding face owner_feature_id")
if selector.get("source") != "runtime_snapshot":
raise ValueError(f"Feature {fid} selector {index} output role must use runtime_snapshot evidence")
if any(selector.get(key) is not None for key in ("stable_id", "snapshot_id", "geometry", "binding_feature_id")):
raise ValueError(f"Feature {fid} selector {index} output role cannot mix stable or geometry evidence")
role_source = selector.get("output_role_source")
if role_source is not None:
source_owner = role_source.get("owner_feature_id") if isinstance(role_source, dict) else None
source_role = role_source.get("output_role") if isinstance(role_source, dict) else None
source_feature = preceding_features.get(str(source_owner or ""))
source_params = (source_feature or {}).get("params") or {}
if not source_owner or source_owner not in feature_ids:
raise ValueError(f"Feature {fid} selector {index} output role source requires a preceding owner_feature_id")
if selector.get("output_role") != "shell.offset_face" or feature.get("atomic_id") != "shell":
raise ValueError(
f"Feature {fid} selector {index} output role source is only supported for shell.offset_face"
)
if source_role not in {"extrude.start", "extrude.end"}:
raise ValueError(
f"Feature {fid} selector {index} shell.offset_face source must be an extrude cap role"
)
if (
source_feature is None
or source_feature.get("atomic_id") != "extrude_add_blind"
or source_params.get("result_mode") != "new_body"
or (source_params.get("end_condition") or {}).get("type") != "blind"
):
raise ValueError(
f"Feature {fid} selector {index} shell.offset_face source requires a preceding direct new_body blind extrusion"
)
elif selector.get("output_role_source") is not None:
raise ValueError(f"Feature {fid} selector {index} output role source requires output_role")
for selector in _mappings(feature):
# ``output_role_source`` is provenance metadata nested inside a
# feature selector, not a selector on its own.
if (
selector.get("output_role") is not None
and selector.get("kind") is not None
and id(selector) not in output_role_selector_ids
):
raise ValueError(f"Feature {fid} output role selectors are only supported in feature.selectors")
if feature.get("atomic_id") == "shell":
target_feature_id = (feature.get("params") or {}).get("target_feature_id")
if target_feature_id is not None and target_feature_id not in feature_ids:
raise ValueError(f"Feature {fid} shell target_feature_id requires a preceding body feature")
if feature.get("atomic_id") == "transform_bodies":
params = feature.get("params") or {}
references = params.get("pattern_instance_refs") or []
for index, reference in enumerate(references):
if not isinstance(reference, dict):
raise ValueError(f"Feature {fid} pattern instance reference {index} is invalid")
pattern_id = str(reference.get("pattern_feature_id") or "")
source_id = str(reference.get("source_feature_id") or "")
pattern = preceding_features.get(pattern_id)
pattern_atomic_id = pattern.get("atomic_id") if pattern is not None else None
if pattern is None or pattern_atomic_id not in {"pattern_circular", "pattern_mirror"}:
raise ValueError(f"Feature {fid} pattern instance reference {index} has a forward or unsupported pattern owner")
if source_id not in {str(value) for value in (pattern.get("params") or {}).get("source_feature_ids") or ()}:
raise ValueError(f"Feature {fid} pattern instance reference {index} names a source outside its pattern")
source = preceding_features.get(source_id)
if (
pattern_atomic_id == "pattern_mirror"
and (source is None or (source.get("params") or {}).get("result_mode") != "new_body")
):
raise ValueError(
f"Feature {fid} pattern instance reference {index} requires a preceding new_body mirror source"
)
instance = reference.get("instance_index")
count = int((pattern.get("params") or {}).get("pattern_count") or 0)
excluded = {int(value) for value in (pattern.get("params") or {}).get("excluded_instance_indices") or ()}
if not isinstance(instance, int) or (
pattern_atomic_id == "pattern_mirror" and instance != 1
) or (
pattern_atomic_id == "pattern_circular" and (instance < 1 or instance >= count or instance in excluded)
):
raise ValueError(f"Feature {fid} pattern instance reference {index} is not a surviving copy")
copy_references = params.get("transform_copy_refs") or []
for index, reference in enumerate(copy_references):
if not isinstance(reference, dict):
raise ValueError(f"Feature {fid} transform COPY reference {index} is invalid")
transform_id = str(reference.get("transform_feature_id") or "")
source_id = str(reference.get("source_feature_id") or "")
transform = preceding_features.get(transform_id)
transform_params = (transform or {}).get("params") or {}
direct_sources = transform_params.get("source_feature_ids") or []
if (
transform is None
or transform.get("atomic_id") != "transform_bodies"
or not bool(transform_params.get("make_copy"))
or not isinstance(direct_sources, list)
or len(direct_sources) < 2
):
raise ValueError(
f"Feature {fid} transform COPY reference {index} requires a preceding multi-source make_copy transform"
)
if source_id not in {str(value) for value in direct_sources}:
raise ValueError(
f"Feature {fid} transform COPY reference {index} names a source outside its transform"
)
for source_id in params.get("source_feature_ids") or []:
source = preceding_features.get(str(source_id))
source_params = (source or {}).get("params") or {}
source_direct_ids = source_params.get("source_feature_ids") or []
if (
source is not None
and source.get("atomic_id") == "transform_bodies"
and bool(source_params.get("make_copy"))
and isinstance(source_direct_ids, list)
and len(source_direct_ids) > 1
):
raise ValueError(
f"Feature {fid} must use transform_copy_refs to select a source of multi-source COPY {source_id}"
)
if feature.get("atomic_id") == "boolean_bodies":
params = feature.get("params") or {}
for parameter in ("target_pattern_instance_refs", "tool_pattern_instance_refs"):
for index, reference in enumerate(params.get(parameter) or []):
if not isinstance(reference, dict):
raise ValueError(f"Feature {fid} {parameter} entry {index} is invalid")
pattern_id = str(reference.get("pattern_feature_id") or "")
source_id = str(reference.get("source_feature_id") or "")
pattern = preceding_features.get(pattern_id)
pattern_atomic_id = pattern.get("atomic_id") if pattern is not None else None
if pattern is None or pattern_atomic_id not in {"pattern_circular", "pattern_mirror"}:
raise ValueError(f"Feature {fid} {parameter} entry {index} has a forward or unsupported pattern owner")
if source_id not in {str(value) for value in (pattern.get("params") or {}).get("source_feature_ids") or ()}:
raise ValueError(f"Feature {fid} {parameter} entry {index} names a source outside its pattern")
source = preceding_features.get(source_id)
if (
pattern_atomic_id == "pattern_mirror"
and (source is None or (source.get("params") or {}).get("result_mode") != "new_body")
):
raise ValueError(
f"Feature {fid} {parameter} entry {index} requires a preceding new_body mirror source"
)
instance = reference.get("instance_index")
count = int((pattern.get("params") or {}).get("pattern_count") or 0)
excluded = {int(value) for value in (pattern.get("params") or {}).get("excluded_instance_indices") or []}
if not isinstance(instance, int) or (
pattern_atomic_id == "pattern_mirror" and instance != 1
) or (
pattern_atomic_id == "pattern_circular" and (instance < 1 or instance >= count or instance in excluded)
):
raise ValueError(f"Feature {fid} {parameter} entry {index} is not a surviving copy")
if feature.get("unresolved"):
unresolved.append({"feature_id": fid, "reasons": list(feature["unresolved"])})
feature_ids.add(fid)
preceding_features[fid] = feature
return {
"schema_version": version,
"feature_count": len(feature_ids),
"sketch_count": len(sketches),
"deferred_feature_ids": deferred,
"unresolved": unresolved,
"future_rebuild_ready": not unresolved,
}