feat(selector): 增加离线候选遍历与严格回放验证 Demo

- 新增 selector_candidate_demo,移除 provenance intent 后枚举候选 selector
- 对候选分支执行有界重建与严格 STEP 比较
- 仅在候选遍历完整且唯一 strict 通过时生成 selector 映射记录
- 增加 selector 候选搜索、预算限制和记录生成的测试
- 保持生产 selector resolver 不受 Demo 逻辑影响
- 更新 CADFS 能力台账,记录 IMPRINT 派生 profile 的 lineage selector 缺口
This commit is contained in:
2026-09-10 15:12:57 +08:00
parent d8fc2a9207
commit 994d06aaea
36 changed files with 9268 additions and 753 deletions
+230 -140
View File
@@ -2,6 +2,7 @@ from __future__ import annotations
import math
from copy import deepcopy
from dataclasses import dataclass
from pathlib import Path
import tempfile
from typing import Any
@@ -95,6 +96,14 @@ def _binding_targets(selector: dict[str, Any]):
yield selector
def _runtime_selector(placeholder: dict[str, Any]) -> dict[str, Any]:
"""Convert a legacy prefix placeholder to the runtime selector contract."""
selector = dict(placeholder)
selector.setdefault("source", "runtime_snapshot")
selector.setdefault("confidence", 1.0)
return selector
def _bound_selector(placeholder: dict[str, Any], records: list[dict[str, Any]]) -> tuple[dict[str, Any], list[dict[str, Any]]]:
bound = []
for record in records:
@@ -109,160 +118,241 @@ def _bound_selector(placeholder: dict[str, Any], records: list[dict[str, Any]])
return bound[0], bound
def _selector_roots(feature: dict[str, Any]) -> list[dict[str, Any]]:
roots = list(feature.get("selectors") or [])
for name in ("end_condition", "reverse_end_condition"):
condition = (feature.get("params") or {}).get(name)
reference = condition.get("reference") if isinstance(condition, dict) else None
if isinstance(reference, dict):
roots.append(reference)
return roots
def _selector_key(selector: dict[str, Any]) -> tuple[Any, ...]:
stable_id = selector.get("stable_id")
if stable_id is not None:
return ("stable_id", str(stable_id))
source = selector.get("output_role_source")
return (
"output_role",
selector.get("owner_feature_id"),
selector.get("kind"),
selector.get("output_role"),
source.get("owner_feature_id") if isinstance(source, dict) else None,
source.get("output_role") if isinstance(source, dict) else None,
)
def _bind_feature_selectors(
feature: dict[str, Any],
*,
registry: Any,
body_id_for_feature: dict[str, str | None],
) -> list[dict[str, Any]]:
"""Bind one feature immediately before it runs in the shared session."""
targets = [target for root in _selector_roots(feature) for target in _binding_targets(root)]
resolved: list[dict[str, Any]] = []
for placeholder in targets:
binding_feature_id = placeholder.get("binding_feature_id")
if binding_feature_id is None:
active_body_id = body_id_for_feature.get("__current__")
else:
if not isinstance(binding_feature_id, str) or binding_feature_id not in body_id_for_feature:
raise ValueError(f"{feature['id']}: selector binding feature is missing or forward")
active_body_id = body_id_for_feature[binding_feature_id]
intent = placeholder.get("selector_intent")
runtime_selector = _runtime_selector(placeholder)
resolution = registry.resolve(runtime_selector, active_body_id=active_body_id)
# Legacy owner-qualified, geometry-free context selectors predate the
# runtime's canonical evidence fields. Their compatibility contract
# permits a unique active context object, but never relaxes a
# provenance, output-role, or instance-locked selector.
if (
resolution.status != "resolved"
and not isinstance(intent, dict)
and not runtime_selector.get("owner_match_required")
and not runtime_selector.get("output_role")
):
fallback = dict(runtime_selector)
fallback.pop("owner_feature_id", None)
resolution = registry.resolve(fallback, active_body_id=active_body_id)
if resolution.status != "resolved":
code = resolution.diagnostic.code if resolution.diagnostic is not None else f"selector_{resolution.status}"
raise ValueError(f"{feature['id']}: {code} during incremental replay")
selected = list(resolution.records or ((resolution.record,) if resolution.record is not None else ()))
if not selected:
raise ValueError(f"{feature['id']}: selector_not_found during incremental replay")
public_records = [record.public_dict() for record in selected]
# Operation-role and provenance selectors stay declarative in bound
# CDSL. A runtime record ID is execution evidence, never their durable
# semantic replacement.
if (isinstance(intent, dict) and intent.get("query_family") != "GEOMETRIC") or placeholder.get("output_role"):
resolved.extend(public_records)
continue
selector, bound_selectors = _bound_selector(placeholder, public_records)
placeholder.clear()
placeholder.update(selector)
resolved.extend(bound_selectors)
selectors = feature.get("selectors") or []
feature["selectors"] = list({_selector_key(selector): selector for selector in selectors}.values())
if feature.get("atomic_id") == "pattern_mirror":
planes = [selector for selector in feature["selectors"] if selector.get("kind") == "plane"]
if len(planes) != 1:
raise ValueError(f"{feature['id']}: mirror plane binding is not unique")
feature.setdefault("params", {})["mirror_plane"] = planes[0]
return resolved
@dataclass
class IncrementalBindingReplay:
"""Bound candidate plus the one session that produced its evidence."""
bound_cdsl: dict[str, Any]
evidence: list[dict[str, Any]]
execution: Any
def bind_and_execute_candidate_selectors(cdsl: dict[str, Any]) -> IncrementalBindingReplay:
"""Bind and execute a CADFS candidate in one ordered kernel replay.
A feature is bound only against topology facts registered by earlier
features in this session. Historical ``binding_feature_id`` values select
a retained snapshot ID, so the binder never needs to re-run a prefix or
recreate an OCC body. Resolver failures retain the live session for the
caller to export its last executable checkpoint.
"""
from engine.cdsl_engine.runtime import prepare_cdsl_execution
bound = deepcopy(cdsl)
execution = prepare_cdsl_execution(bound)
if not execution.analysis.runtime_eligible:
first = next((result for result in execution.analysis.feature_results if not result.executable), None)
code = first.blockers[0].code if first and first.blockers else "runtime_ineligible"
raise ValueError(code)
bound_features = {str(feature.get("id") or ""): feature for feature in bound.get("features") or []}
body_id_for_feature: dict[str, str | None] = {"__current__": None}
evidence: list[dict[str, Any]] = []
for index, node in enumerate(execution.analysis.plan):
feature = node.source_feature
try:
resolved = _bind_feature_selectors(
feature,
registry=execution.session.topology,
body_id_for_feature=body_id_for_feature,
)
public_feature = bound_features[node.feature_id]
public_feature["selectors"] = deepcopy(feature.get("selectors") or [])
public_feature["params"] = deepcopy(feature.get("params") or {})
evidence.append({
"feature_id": node.feature_id,
"prefix_feature_count": index,
"selectors": public_feature["selectors"],
"resolved": resolved,
})
execution.execute_next(strict=True)
body_id_for_feature[node.feature_id] = execution.session.body_id
body_id_for_feature["__current__"] = execution.session.body_id
except Exception as error:
setattr(error, "bound_cdsl", bound)
setattr(error, "selector_binding", evidence)
setattr(error, "incremental_execution", execution)
setattr(error, "failed_feature_id", node.feature_id)
raise
return IncrementalBindingReplay(bound, evidence, execution)
def bind_candidate_selectors(cdsl: dict[str, Any]) -> tuple[dict[str, Any], list[dict[str, Any]]]:
"""Rebuild every selector-bearing prefix and bind against its active body."""
"""Legacy diagnostic binder for incomplete or externally supplied CDSL.
The production CADFS rebuild path uses ``bind_and_execute_candidate_selectors``.
This compatibility entry point deliberately keeps prefix replay explicit
for callers that need to inspect a partial document before it is runtime
eligible. It rehydrates exact exported topology facts and calls the same
resolver; provenance selectors never become geometry guesses here.
"""
from engine.cdsl_engine.runtime import rebuild_cdsl
bound = deepcopy(cdsl); evidence = []
from engine.cdsl_engine.topology import TopologyRegistry
bound = deepcopy(cdsl)
evidence: list[dict[str, Any]] = []
with tempfile.TemporaryDirectory(prefix="cadfs-bind-") as temporary:
for index, feature in enumerate(bound.get("features") or []):
roots = list(feature.get("selectors") or [])
for name in ("end_condition", "reverse_end_condition"):
condition = (feature.get("params") or {}).get(name)
reference = condition.get("reference") if isinstance(condition, dict) else None
if isinstance(reference, dict): roots.append(reference)
targets = [target for root in roots for target in _binding_targets(root)]
if not targets: continue
prefix_cache: dict[int, tuple[list[dict[str, Any]], str | None]] = {}
targets = [target for root in _selector_roots(feature) for target in _binding_targets(root)]
if not targets:
continue
prefix_cache: dict[int, tuple[list[dict[str, Any]], str | None, list[dict[str, Any]]]] = {}
def prefix_records(binding_feature_id: str | None, owner_feature_id: str | None) -> list[dict[str, Any]]:
def prefix_snapshot(binding_feature_id: str | None) -> tuple[list[dict[str, Any]], str | None, list[dict[str, Any]]]:
prefix_count = index
if binding_feature_id is not None:
binding_index = next((item_index for item_index, item in enumerate(bound["features"][:index]) if item["id"] == binding_feature_id), None)
if binding_index is None: raise ValueError(f"{feature['id']}: selector binding feature is missing or forward")
binding_index = next(
(item_index for item_index, item in enumerate(bound["features"][:index]) if item["id"] == binding_feature_id),
None,
)
if binding_index is None:
raise ValueError(f"{feature['id']}: selector binding feature is missing or forward")
prefix_count = binding_index + 1
if prefix_count not in prefix_cache:
prefix = deepcopy(bound); prefix["features"] = bound["features"][:prefix_count]
if not prefix["features"]: raise ValueError(f"{feature['id']}: selector has no executable prefix")
report = rebuild_cdsl(prefix, Path(temporary) / f"prefix-{index}-{prefix_count}.step", strict=True)
body_id = next((item.get("body_id") for item in reversed(report.get("feature_results") or []) if item.get("body_id")), None)
prefix_cache[prefix_count] = (list(report.get("topology_records") or []), body_id)
all_records, body_id = prefix_cache[prefix_count]
return [
item for item in all_records
# Reference planes and axes are session context, not body
# topology. They must remain available while binding a mirror
# or extent selector against a body-bearing prefix.
if item.get("kind") in {"plane", "axis"}
or not body_id
or item.get("body_id") == body_id
or str(item.get("body_id") or "").startswith(f"{body_id}:")
]
resolved = []
for placeholder in targets:
records = prefix_records(placeholder.get("binding_feature_id"), placeholder.get("owner_feature_id"))
output_role = str(placeholder.get("output_role") or "").strip()
intent = placeholder.get("selector_intent")
if (
isinstance(intent, dict)
and intent.get("query_family") != "GEOMETRIC"
and not output_role
):
# Prefix rebuilding is retained as a diagnostic adapter.
# It must not turn an unproven FeatureScript provenance
# query into a geometry-scored stable selector.
raise ValueError(
f"{feature['id']}: selector_kernel_history_missing after prefix rebuild"
prefix = deepcopy(bound)
prefix["features"] = bound["features"][:prefix_count]
if not prefix["features"]:
raise ValueError(f"{feature['id']}: selector has no executable prefix")
report = rebuild_cdsl(
prefix,
Path(temporary) / f"prefix-{index}-{prefix_count}.step",
strict=True,
)
if output_role:
# Builder output roles are not geometry placeholders. They
# remain in the bound CDSL so runtime can resolve the
# current active B-rep face through exact kernel history.
# Replacing one with stable_id/geometry would mix evidence
# and make a stale snapshot appear durable.
owner = placeholder.get("owner_feature_id")
role_source = placeholder.get("output_role_source")
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
candidates = [
record for record in records
if record.get("kind") == placeholder.get("kind")
and owner in (record.get("owner_feature_ids") or [record.get("feature_id")])
and output_role in (record.get("output_roles") or [])
and (
role_source is None
or any(
item.get("output_role") == output_role
and item.get("owner_feature_id") == source_owner
and item.get("source_output_role") == source_role
for item in record.get("output_role_sources") or []
)
)
]
if len(candidates) != 1:
status = "not_found" if not candidates else "ambiguous"
raise ValueError(
f"{feature['id']}: selector_output_role_{status} after prefix rebuild"
)
resolved.append(candidates[0])
continue
geometry = placeholder.get("geometry") or {}
candidates = _circle_records(geometry, records) if geometry.get("source_circle_radius_mm") else []
if not candidates:
same_kind = [record for record in records if record.get("kind") == placeholder.get("kind")]
owner = placeholder.get("owner_feature_id")
owner_matches = [record for record in same_kind if owner in (record.get("owner_feature_ids") or [record.get("feature_id")])]
# Pattern instance provenance is stronger than an ordinary
# feature owner: an unresolved instance cannot fall back to
# an aggregate face with matching geometry. Ordinary source
# selectors retain their established geometry-binding path
# across topology-changing dress-up operations.
pool = owner_matches if placeholder.get("owner_match_required") else (owner_matches or same_kind)
if not geometry:
# Context selectors (notably a generated mirror plane)
# may have no geometric snapshot. Their owner-qualified
# singleton identity is sufficient and must not be scored
# as a zero-information geometric match.
if len(pool) != 1:
status = "not_found" if not pool else "ambiguous"
raise ValueError(f"{feature['id']}: selector_{status} after prefix rebuild")
candidates = [pool[0]]
else:
scored = [(score, record) for record in pool if (score := _score(geometry, record.get("geometry") or {})) is not None and score >= 0.8]
# A source owner is preferred for ordinary geometry
# selectors, but can be retained by an unrelated exact
# continuation after a dress-up. If none of that
# owner's active records satisfies the full geometry
# signature, bind against the current active body and
# still require a unique threshold-qualified match.
# Instance-qualified selectors never take this path.
if not scored and not placeholder.get("owner_match_required"):
scored = [
(score, record)
for record in same_kind
if (score := _score(geometry, record.get("geometry") or {})) is not None
and score >= 0.8
]
scored.sort(key=lambda value: (-value[0], str(value[1].get("record_id"))))
if scored:
if placeholder.get("match_mode") != "all" and len(scored) > 1 and abs(scored[0][0] - scored[1][0]) <= 1e-9:
raise ValueError(f"{feature['id']}: selector_ambiguous after prefix rebuild")
candidates = [item[1] for item in scored] if placeholder.get("match_mode") == "all" else [scored[0][1]]
if not candidates: raise ValueError(f"{feature['id']}: selector_not_found after prefix rebuild")
selector, bound_selectors = _bound_selector(placeholder, candidates)
placeholder.clear(); placeholder.update(selector)
resolved.extend(bound_selectors)
feature_selectors = feature.get("selectors") or []
def selector_key(selector: dict[str, Any]) -> tuple[Any, ...]:
stable_id = selector.get("stable_id")
if stable_id is not None:
return ("stable_id", str(stable_id))
source = selector.get("output_role_source")
return (
"output_role",
selector.get("owner_feature_id"),
selector.get("kind"),
selector.get("output_role"),
source.get("owner_feature_id") if isinstance(source, dict) else None,
source.get("output_role") if isinstance(source, dict) else None,
)
body_id = next(
(item.get("body_id") for item in reversed(report.get("feature_results") or []) if item.get("body_id")),
None,
)
prefix_cache[prefix_count] = (
list(report.get("topology_records") or []),
body_id,
list(report.get("topology_deltas") or []),
)
return prefix_cache[prefix_count]
unique = {selector_key(selector): selector for selector in feature_selectors}
feature["selectors"] = list(unique.values())
resolved: list[dict[str, Any]] = []
for placeholder in targets:
records, body_id, topology_deltas = prefix_snapshot(placeholder.get("binding_feature_id"))
registry = TopologyRegistry.from_public_snapshot(records, topology_deltas)
intent = placeholder.get("selector_intent")
runtime_selector = _runtime_selector(placeholder)
resolution = registry.resolve(runtime_selector, active_body_id=body_id)
if (
resolution.status != "resolved"
and not isinstance(intent, dict)
and not runtime_selector.get("owner_match_required")
and not runtime_selector.get("output_role")
):
fallback = dict(runtime_selector)
fallback.pop("owner_feature_id", None)
resolution = registry.resolve(fallback, active_body_id=body_id)
if resolution.status != "resolved":
code = resolution.diagnostic.code if resolution.diagnostic is not None else f"selector_{resolution.status}"
raise ValueError(f"{feature['id']}: {code} after prefix rebuild")
selected = list(resolution.records or ((resolution.record,) if resolution.record is not None else ()))
if not selected:
raise ValueError(f"{feature['id']}: selector_not_found after prefix rebuild")
public_records = [record.public_dict() for record in selected]
if (isinstance(intent, dict) and intent.get("query_family") != "GEOMETRIC") or placeholder.get("output_role"):
resolved.extend(public_records)
continue
selector, bound_selectors = _bound_selector(placeholder, public_records)
placeholder.clear()
placeholder.update(selector)
resolved.extend(bound_selectors)
feature["selectors"] = list({_selector_key(selector): selector for selector in feature.get("selectors") or []}.values())
if feature.get("atomic_id") == "pattern_mirror":
planes = [selector for selector in feature["selectors"] if selector.get("kind") == "plane"]
if len(planes) != 1:
raise ValueError(f"{feature['id']}: mirror plane binding is not unique")
feature.setdefault("params", {})["mirror_plane"] = planes[0]
evidence.append({"feature_id": feature["id"], "prefix_feature_count": index, "selectors": feature["selectors"], "resolved": resolved})
evidence.append({
"feature_id": feature["id"],
"prefix_feature_count": index,
"selectors": feature["selectors"],
"resolved": resolved,
})
return bound, evidence