refactor: extract store constants and remaining domain helpers
Move shared constants + _bounded_int_env into store_constants (a leaf module), and the remaining module-level domain helpers (validation, query signatures, search-hint evidence, watchlist selection, knowledge type/provenance) into store_serialization. sqlite_store.py is now the CopyrighterStore class plus thin imports: 3613 -> 3368 lines (5333 -> 3368 overall, -37%). All behavior-preserving.
This commit is contained in:
유창욱
parent
8e53139029
commit
3be7b016ce
3 changed files with 314 additions and 273 deletions
|
|
@ -28,7 +28,6 @@ from rights_filter.domain.records import (
|
|||
EvidenceSource,
|
||||
InMemoryRightsFilterRepository,
|
||||
KnowledgeBaseEntry,
|
||||
KnowledgeEntryType,
|
||||
)
|
||||
from rights_filter.integrations.cloud_vision_web_detection import (
|
||||
CloudVisionWebDetectionAdapter,
|
||||
|
|
@ -38,6 +37,18 @@ from rights_filter.integrations.env_clients import ProviderRuntime, build_provid
|
|||
from rights_filter.integrations.external_policy import ExternalApiPolicy
|
||||
from rights_filter.jobs.batch_analyzer import BatchAnalyzer, SubmissionImage
|
||||
from rights_filter.server.image_store import LocalSubmissionImageStore, SUPPORTED_IMAGE_SUFFIXES
|
||||
from rights_filter.server.store_constants import (
|
||||
DEFAULT_COVERAGE_GOOD_THRESHOLD,
|
||||
DEFAULT_COVERAGE_WARN_THRESHOLD,
|
||||
DEFAULT_FACE_CROP_RETENTION_DAYS,
|
||||
DEFAULT_QUERY_COVERAGE_GOOD_THRESHOLD,
|
||||
DEFAULT_QUERY_COVERAGE_WARN_THRESHOLD,
|
||||
EVIDENCE_OPERATOR_STATUSES,
|
||||
MAX_COVERAGE_THRESHOLD,
|
||||
MIN_COVERAGE_THRESHOLD,
|
||||
NON_CONTRIBUTING_OPERATOR_STATUSES,
|
||||
_bounded_int_env,
|
||||
)
|
||||
from rights_filter.server.store_remote_fetch import (
|
||||
_fetch_page_url_bytes,
|
||||
_fetch_stylesheet_url_bytes,
|
||||
|
|
@ -58,27 +69,42 @@ from rights_filter.server.store_schema import (
|
|||
_ensure_typed_columns,
|
||||
)
|
||||
from rights_filter.server.store_serialization import (
|
||||
_default_evidence_contribution,
|
||||
_domain_evidence_from_ui,
|
||||
_evidence_id,
|
||||
_evidence_matches_provider,
|
||||
_evidence_payload,
|
||||
_existing_google_custom_query_signatures,
|
||||
_existing_naver_query_signatures,
|
||||
_external_provider_ids,
|
||||
_external_provider_state_for_submission,
|
||||
_face_crop_web_evidence,
|
||||
_google_custom_image_query_signature,
|
||||
_google_custom_web_query_signature,
|
||||
_google_weak_label_title,
|
||||
_image_size_from_bytes,
|
||||
_image_suffix_from_url,
|
||||
_is_google_weak_label_payload,
|
||||
_knowledge_entry_type,
|
||||
_knowledge_provenance,
|
||||
_knowledge_type_value,
|
||||
_naver_blog_query_signature,
|
||||
_naver_query_signature,
|
||||
_naver_web_query_signature,
|
||||
_now_label,
|
||||
_provider_item_failed,
|
||||
_provider_item_has_result,
|
||||
_query_history_status,
|
||||
_safe_filename,
|
||||
_safe_image_suffix,
|
||||
_stable_id,
|
||||
_strip_html,
|
||||
_submission_payload,
|
||||
_submission_search_hint_evidence,
|
||||
_timestamp_id,
|
||||
_validate_payload,
|
||||
_validate_table,
|
||||
_watchlist_source_evidence,
|
||||
)
|
||||
from rights_filter.server.store_text import _text_list, _unique_texts
|
||||
from rights_filter.server.store_url_utils import (
|
||||
|
|
@ -90,116 +116,6 @@ from rights_filter.server.store_url_utils import (
|
|||
)
|
||||
|
||||
|
||||
EVIDENCE_OPERATOR_STATUSES = {
|
||||
"used_for_judgment": "판단에 사용",
|
||||
"irrelevant": "무관",
|
||||
"false_positive": "오탐",
|
||||
"pending": "보류",
|
||||
}
|
||||
NON_CONTRIBUTING_OPERATOR_STATUSES = {"irrelevant", "false_positive"}
|
||||
SUBMISSION_FINAL_STATUSES = {"approved", "rejected", "corrected"}
|
||||
QUEUE_ID_PLACEHOLDER = ""
|
||||
STORE_TABLES = {
|
||||
"submissions",
|
||||
"evidence",
|
||||
"providers",
|
||||
"knowledge_entries",
|
||||
"collection_candidates",
|
||||
"corrections",
|
||||
"audit_events",
|
||||
"submission_queues",
|
||||
}
|
||||
RISK_BANDS = ("low", "medium", "high", "failed", "pending")
|
||||
DECISION_STATUSES = ("unreviewed", "held", "rejected", "approved", "corrected")
|
||||
EVIDENCE_STATUSES = (
|
||||
"active",
|
||||
"auto",
|
||||
"manual",
|
||||
"queued",
|
||||
"rerun",
|
||||
"weak",
|
||||
"used_for_judgment",
|
||||
"irrelevant",
|
||||
"false_positive",
|
||||
"pending",
|
||||
)
|
||||
KNOWLEDGE_STATUSES = ("confirmed", "watchlist", "excluded")
|
||||
COLLECTION_STATUSES = ("candidate", "promoted")
|
||||
DEFAULT_COVERAGE_GOOD_THRESHOLD = 70
|
||||
DEFAULT_COVERAGE_WARN_THRESHOLD = 40
|
||||
DEFAULT_QUERY_COVERAGE_GOOD_THRESHOLD = 70
|
||||
DEFAULT_QUERY_COVERAGE_WARN_THRESHOLD = 40
|
||||
MIN_COVERAGE_THRESHOLD = 0
|
||||
MAX_COVERAGE_THRESHOLD = 100
|
||||
DEFAULT_FACE_CROP_RETENTION_DAYS = 90
|
||||
PAYLOAD_REQUIRED_FIELDS = {
|
||||
"submissions": {
|
||||
"id",
|
||||
"title",
|
||||
"asset",
|
||||
"riskScore",
|
||||
"riskBand",
|
||||
"decisionStatus",
|
||||
"providerState",
|
||||
"fileFacts",
|
||||
"evidence",
|
||||
},
|
||||
"evidence": {
|
||||
"id",
|
||||
"source",
|
||||
"title",
|
||||
"confidence",
|
||||
"status",
|
||||
"submission_id",
|
||||
},
|
||||
"providers": {
|
||||
"id",
|
||||
"name",
|
||||
"enabled",
|
||||
"usage",
|
||||
"quota",
|
||||
"lastSuccess",
|
||||
"lastFailure",
|
||||
},
|
||||
"knowledge_entries": {
|
||||
"id",
|
||||
"name",
|
||||
"type",
|
||||
"provenance",
|
||||
"active",
|
||||
"entryStatus",
|
||||
"sampleFingerprints",
|
||||
},
|
||||
"collection_candidates": {
|
||||
"id",
|
||||
"provider",
|
||||
"query",
|
||||
"title",
|
||||
"status",
|
||||
"sourceUrl",
|
||||
"collectedEpoch",
|
||||
"sampleFingerprints",
|
||||
},
|
||||
"corrections": {
|
||||
"id",
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
|
||||
|
||||
def _bounded_int_env(value: str | None, default: int, minimum: int, maximum: int) -> int:
|
||||
try:
|
||||
parsed = int(value) if value is not None else default
|
||||
except (TypeError, ValueError):
|
||||
return default
|
||||
if parsed < minimum:
|
||||
return minimum
|
||||
if parsed > maximum:
|
||||
return maximum
|
||||
return parsed
|
||||
|
||||
|
||||
class CopyrighterStore:
|
||||
def __init__(
|
||||
self,
|
||||
|
|
@ -3450,164 +3366,3 @@ class CopyrighterStore:
|
|||
return grouped
|
||||
|
||||
|
||||
def _validate_table(table: str) -> None:
|
||||
if table not in STORE_TABLES:
|
||||
raise ValueError(f"unsupported store table: {table}")
|
||||
|
||||
|
||||
def _validate_payload(table: str, id_value: str, payload: dict[str, Any]) -> None:
|
||||
_validate_table(table)
|
||||
if not isinstance(payload, dict):
|
||||
raise ValueError(f"{table} payload must be an object")
|
||||
|
||||
required = PAYLOAD_REQUIRED_FIELDS.get(table, set())
|
||||
missing = sorted(field for field in required if field not in payload)
|
||||
if missing:
|
||||
raise ValueError(f"{table} payload missing required fields: {', '.join(missing)}")
|
||||
|
||||
if table != "audit_events" and "id" in payload and str(payload["id"]) != str(id_value):
|
||||
raise ValueError(f"{table} payload id does not match storage id")
|
||||
|
||||
|
||||
def _existing_naver_query_signatures(evidence: list[dict[str, Any]]) -> set[str]:
|
||||
return {
|
||||
str(item.get("querySignature") or _naver_query_signature(str(item.get("query", ""))))
|
||||
for item in evidence
|
||||
if item.get("query") and item.get("source") in {"naver", "failure"}
|
||||
}
|
||||
|
||||
|
||||
def _existing_google_custom_query_signatures(evidence: list[dict[str, Any]]) -> set[str]:
|
||||
return {
|
||||
str(item.get("querySignature") or "")
|
||||
for item in evidence
|
||||
if item.get("query")
|
||||
and item.get("source") in {"google", "failure"}
|
||||
and str(item.get("domain", "")) == "google_custom_search"
|
||||
and item.get("querySignature")
|
||||
}
|
||||
|
||||
|
||||
def _google_custom_image_query_signature(query: str) -> str:
|
||||
return "google-custom-image:" + " ".join(query.lower().split())
|
||||
|
||||
|
||||
def _google_custom_web_query_signature(query: str) -> str:
|
||||
return "google-custom-web:" + " ".join(query.lower().split())
|
||||
|
||||
|
||||
def _naver_query_signature(query: str) -> str:
|
||||
return "naver:" + " ".join(query.lower().split())
|
||||
|
||||
|
||||
def _naver_blog_query_signature(query: str) -> str:
|
||||
return "naver-blog:" + " ".join(query.lower().split())
|
||||
|
||||
|
||||
def _naver_web_query_signature(query: str) -> str:
|
||||
return "naver-web:" + " ".join(query.lower().split())
|
||||
|
||||
|
||||
def _submission_search_hint_evidence(record: dict[str, Any]) -> list[Evidence]:
|
||||
hints: list[Evidence] = []
|
||||
title = str(record.get("title", "")).strip()
|
||||
if title:
|
||||
hints.append(_local_query_hint_evidence(title, "title"))
|
||||
|
||||
file_value = str(record.get("file", record.get("asset", ""))).strip()
|
||||
file_stem = Path(urlparse(file_value).path).stem if file_value else ""
|
||||
if file_stem and file_stem != title:
|
||||
hints.append(_local_query_hint_evidence(file_stem, "file"))
|
||||
return hints
|
||||
|
||||
|
||||
def _local_query_hint_evidence(query: str, hint_source: str) -> Evidence:
|
||||
return Evidence(
|
||||
source=EvidenceSource.FINGERPRINT,
|
||||
reason="Local submission search hint",
|
||||
confidence=0.0,
|
||||
data={
|
||||
"local_query_hint": True,
|
||||
"query": query,
|
||||
"hint_source": hint_source,
|
||||
},
|
||||
)
|
||||
|
||||
|
||||
def _query_history_status(evidence: list[Evidence]) -> str:
|
||||
if any(item.source == EvidenceSource.ENRICHMENT_FAILURE for item in evidence):
|
||||
return "failed"
|
||||
if evidence and all(item.source == EvidenceSource.SEARCH_SKIPPED for item in evidence):
|
||||
return "skipped"
|
||||
return "auto"
|
||||
|
||||
|
||||
def _default_evidence_contribution(payload: dict[str, Any]) -> bool:
|
||||
source = str(payload.get("source", ""))
|
||||
if source in {"llm", "failure"}:
|
||||
return False
|
||||
if bool(payload.get("faceCropSearch", False)):
|
||||
return False
|
||||
if _is_google_weak_label_payload(payload):
|
||||
return False
|
||||
return True
|
||||
|
||||
|
||||
def _watchlist_source_evidence(evidence: list[dict[str, Any]]) -> list[dict[str, Any]]:
|
||||
used = [
|
||||
item
|
||||
for item in evidence
|
||||
if item.get("operatorStatus") == "used_for_judgment"
|
||||
and item.get("id")
|
||||
]
|
||||
if used:
|
||||
return _sorted_evidence_for_watchlist(used)[:5]
|
||||
|
||||
candidates = [
|
||||
item
|
||||
for item in evidence
|
||||
if item.get("id")
|
||||
and item.get("operatorStatus") not in NON_CONTRIBUTING_OPERATOR_STATUSES
|
||||
and item.get("source") not in {"llm", "failure"}
|
||||
and item.get("title") != "Image fingerprints generated"
|
||||
]
|
||||
if candidates:
|
||||
return _sorted_evidence_for_watchlist(candidates)[:5]
|
||||
|
||||
fallback = [
|
||||
item
|
||||
for item in evidence
|
||||
if item.get("id")
|
||||
and item.get("operatorStatus") not in NON_CONTRIBUTING_OPERATOR_STATUSES
|
||||
]
|
||||
return _sorted_evidence_for_watchlist(fallback)[:5]
|
||||
|
||||
|
||||
def _sorted_evidence_for_watchlist(evidence: list[dict[str, Any]]) -> list[dict[str, Any]]:
|
||||
return sorted(
|
||||
evidence,
|
||||
key=lambda item: (
|
||||
1 if item.get("contributed", True) else 0,
|
||||
float(item.get("confidence", 0) or 0),
|
||||
str(item.get("retrievedAt", "")),
|
||||
),
|
||||
reverse=True,
|
||||
)
|
||||
|
||||
|
||||
def _knowledge_type_value(value: str) -> str:
|
||||
normalized = value.strip() or "other"
|
||||
aliases = {
|
||||
"public_figure": "celebrity",
|
||||
"rejected_reference": "rejected_image",
|
||||
}
|
||||
return aliases.get(normalized, normalized)
|
||||
|
||||
|
||||
def _knowledge_entry_type(value: str) -> KnowledgeEntryType:
|
||||
try:
|
||||
return KnowledgeEntryType(_knowledge_type_value(value))
|
||||
except ValueError:
|
||||
return KnowledgeEntryType.OTHER
|
||||
|
||||
|
||||
|
|
|
|||
113
src/rights_filter/server/store_constants.py
Normal file
113
src/rights_filter/server/store_constants.py
Normal file
|
|
@ -0,0 +1,113 @@
|
|||
"""Shared constants and small config helpers for the SQLite store and its
|
||||
extracted submodules. Leaf module (no internal imports) so any store module can
|
||||
depend on it without import cycles. Extracted from sqlite_store.py unchanged.
|
||||
"""
|
||||
|
||||
from __future__ import annotations
|
||||
|
||||
EVIDENCE_OPERATOR_STATUSES = {
|
||||
"used_for_judgment": "판단에 사용",
|
||||
"irrelevant": "무관",
|
||||
"false_positive": "오탐",
|
||||
"pending": "보류",
|
||||
}
|
||||
NON_CONTRIBUTING_OPERATOR_STATUSES = {"irrelevant", "false_positive"}
|
||||
SUBMISSION_FINAL_STATUSES = {"approved", "rejected", "corrected"}
|
||||
QUEUE_ID_PLACEHOLDER = ""
|
||||
STORE_TABLES = {
|
||||
"submissions",
|
||||
"evidence",
|
||||
"providers",
|
||||
"knowledge_entries",
|
||||
"collection_candidates",
|
||||
"corrections",
|
||||
"audit_events",
|
||||
"submission_queues",
|
||||
}
|
||||
RISK_BANDS = ("low", "medium", "high", "failed", "pending")
|
||||
DECISION_STATUSES = ("unreviewed", "held", "rejected", "approved", "corrected")
|
||||
EVIDENCE_STATUSES = (
|
||||
"active",
|
||||
"auto",
|
||||
"manual",
|
||||
"queued",
|
||||
"rerun",
|
||||
"weak",
|
||||
"used_for_judgment",
|
||||
"irrelevant",
|
||||
"false_positive",
|
||||
"pending",
|
||||
)
|
||||
KNOWLEDGE_STATUSES = ("confirmed", "watchlist", "excluded")
|
||||
COLLECTION_STATUSES = ("candidate", "promoted")
|
||||
DEFAULT_COVERAGE_GOOD_THRESHOLD = 70
|
||||
DEFAULT_COVERAGE_WARN_THRESHOLD = 40
|
||||
DEFAULT_QUERY_COVERAGE_GOOD_THRESHOLD = 70
|
||||
DEFAULT_QUERY_COVERAGE_WARN_THRESHOLD = 40
|
||||
MIN_COVERAGE_THRESHOLD = 0
|
||||
MAX_COVERAGE_THRESHOLD = 100
|
||||
DEFAULT_FACE_CROP_RETENTION_DAYS = 90
|
||||
PAYLOAD_REQUIRED_FIELDS = {
|
||||
"submissions": {
|
||||
"id",
|
||||
"title",
|
||||
"asset",
|
||||
"riskScore",
|
||||
"riskBand",
|
||||
"decisionStatus",
|
||||
"providerState",
|
||||
"fileFacts",
|
||||
"evidence",
|
||||
},
|
||||
"evidence": {
|
||||
"id",
|
||||
"source",
|
||||
"title",
|
||||
"confidence",
|
||||
"status",
|
||||
"submission_id",
|
||||
},
|
||||
"providers": {
|
||||
"id",
|
||||
"name",
|
||||
"enabled",
|
||||
"usage",
|
||||
"quota",
|
||||
"lastSuccess",
|
||||
"lastFailure",
|
||||
},
|
||||
"knowledge_entries": {
|
||||
"id",
|
||||
"name",
|
||||
"type",
|
||||
"provenance",
|
||||
"active",
|
||||
"entryStatus",
|
||||
"sampleFingerprints",
|
||||
},
|
||||
"collection_candidates": {
|
||||
"id",
|
||||
"provider",
|
||||
"query",
|
||||
"title",
|
||||
"status",
|
||||
"sourceUrl",
|
||||
"collectedEpoch",
|
||||
"sampleFingerprints",
|
||||
},
|
||||
"corrections": {
|
||||
"id",
|
||||
},
|
||||
}
|
||||
|
||||
|
||||
def _bounded_int_env(value: str | None, default: int, minimum: int, maximum: int) -> int:
|
||||
try:
|
||||
parsed = int(value) if value is not None else default
|
||||
except (TypeError, ValueError):
|
||||
return default
|
||||
if parsed < minimum:
|
||||
return minimum
|
||||
if parsed > maximum:
|
||||
return maximum
|
||||
return parsed
|
||||
|
|
@ -18,8 +18,18 @@ from pathlib import Path
|
|||
from typing import Any
|
||||
from urllib.parse import urlparse
|
||||
|
||||
from rights_filter.domain.records import Evidence, EvidenceSource, KnowledgeProvenance
|
||||
from rights_filter.domain.records import (
|
||||
Evidence,
|
||||
EvidenceSource,
|
||||
KnowledgeEntryType,
|
||||
KnowledgeProvenance,
|
||||
)
|
||||
from rights_filter.server.image_store import SUPPORTED_IMAGE_SUFFIXES
|
||||
from rights_filter.server.store_constants import (
|
||||
NON_CONTRIBUTING_OPERATOR_STATUSES,
|
||||
PAYLOAD_REQUIRED_FIELDS,
|
||||
STORE_TABLES,
|
||||
)
|
||||
from rights_filter.server.store_text import _text_list, _unique_texts
|
||||
|
||||
|
||||
|
|
@ -422,3 +432,166 @@ def _label_to_epoch(value: str) -> int:
|
|||
|
||||
def _timestamp_id() -> str:
|
||||
return datetime.now().strftime("%Y%m%d%H%M%S%f")
|
||||
|
||||
|
||||
def _validate_table(table: str) -> None:
|
||||
if table not in STORE_TABLES:
|
||||
raise ValueError(f"unsupported store table: {table}")
|
||||
|
||||
|
||||
def _validate_payload(table: str, id_value: str, payload: dict[str, Any]) -> None:
|
||||
_validate_table(table)
|
||||
if not isinstance(payload, dict):
|
||||
raise ValueError(f"{table} payload must be an object")
|
||||
|
||||
required = PAYLOAD_REQUIRED_FIELDS.get(table, set())
|
||||
missing = sorted(field for field in required if field not in payload)
|
||||
if missing:
|
||||
raise ValueError(f"{table} payload missing required fields: {', '.join(missing)}")
|
||||
|
||||
if table != "audit_events" and "id" in payload and str(payload["id"]) != str(id_value):
|
||||
raise ValueError(f"{table} payload id does not match storage id")
|
||||
|
||||
|
||||
def _existing_naver_query_signatures(evidence: list[dict[str, Any]]) -> set[str]:
|
||||
return {
|
||||
str(item.get("querySignature") or _naver_query_signature(str(item.get("query", ""))))
|
||||
for item in evidence
|
||||
if item.get("query") and item.get("source") in {"naver", "failure"}
|
||||
}
|
||||
|
||||
|
||||
def _existing_google_custom_query_signatures(evidence: list[dict[str, Any]]) -> set[str]:
|
||||
return {
|
||||
str(item.get("querySignature") or "")
|
||||
for item in evidence
|
||||
if item.get("query")
|
||||
and item.get("source") in {"google", "failure"}
|
||||
and str(item.get("domain", "")) == "google_custom_search"
|
||||
and item.get("querySignature")
|
||||
}
|
||||
|
||||
|
||||
def _google_custom_image_query_signature(query: str) -> str:
|
||||
return "google-custom-image:" + " ".join(query.lower().split())
|
||||
|
||||
|
||||
def _google_custom_web_query_signature(query: str) -> str:
|
||||
return "google-custom-web:" + " ".join(query.lower().split())
|
||||
|
||||
|
||||
def _naver_query_signature(query: str) -> str:
|
||||
return "naver:" + " ".join(query.lower().split())
|
||||
|
||||
|
||||
def _naver_blog_query_signature(query: str) -> str:
|
||||
return "naver-blog:" + " ".join(query.lower().split())
|
||||
|
||||
|
||||
def _naver_web_query_signature(query: str) -> str:
|
||||
return "naver-web:" + " ".join(query.lower().split())
|
||||
|
||||
|
||||
def _submission_search_hint_evidence(record: dict[str, Any]) -> list[Evidence]:
|
||||
hints: list[Evidence] = []
|
||||
title = str(record.get("title", "")).strip()
|
||||
if title:
|
||||
hints.append(_local_query_hint_evidence(title, "title"))
|
||||
|
||||
file_value = str(record.get("file", record.get("asset", ""))).strip()
|
||||
file_stem = Path(urlparse(file_value).path).stem if file_value else ""
|
||||
if file_stem and file_stem != title:
|
||||
hints.append(_local_query_hint_evidence(file_stem, "file"))
|
||||
return hints
|
||||
|
||||
|
||||
def _local_query_hint_evidence(query: str, hint_source: str) -> Evidence:
|
||||
return Evidence(
|
||||
source=EvidenceSource.FINGERPRINT,
|
||||
reason="Local submission search hint",
|
||||
confidence=0.0,
|
||||
data={
|
||||
"local_query_hint": True,
|
||||
"query": query,
|
||||
"hint_source": hint_source,
|
||||
},
|
||||
)
|
||||
|
||||
|
||||
def _query_history_status(evidence: list[Evidence]) -> str:
|
||||
if any(item.source == EvidenceSource.ENRICHMENT_FAILURE for item in evidence):
|
||||
return "failed"
|
||||
if evidence and all(item.source == EvidenceSource.SEARCH_SKIPPED for item in evidence):
|
||||
return "skipped"
|
||||
return "auto"
|
||||
|
||||
|
||||
def _default_evidence_contribution(payload: dict[str, Any]) -> bool:
|
||||
source = str(payload.get("source", ""))
|
||||
if source in {"llm", "failure"}:
|
||||
return False
|
||||
if bool(payload.get("faceCropSearch", False)):
|
||||
return False
|
||||
if _is_google_weak_label_payload(payload):
|
||||
return False
|
||||
return True
|
||||
|
||||
|
||||
def _watchlist_source_evidence(evidence: list[dict[str, Any]]) -> list[dict[str, Any]]:
|
||||
used = [
|
||||
item
|
||||
for item in evidence
|
||||
if item.get("operatorStatus") == "used_for_judgment"
|
||||
and item.get("id")
|
||||
]
|
||||
if used:
|
||||
return _sorted_evidence_for_watchlist(used)[:5]
|
||||
|
||||
candidates = [
|
||||
item
|
||||
for item in evidence
|
||||
if item.get("id")
|
||||
and item.get("operatorStatus") not in NON_CONTRIBUTING_OPERATOR_STATUSES
|
||||
and item.get("source") not in {"llm", "failure"}
|
||||
and item.get("title") != "Image fingerprints generated"
|
||||
]
|
||||
if candidates:
|
||||
return _sorted_evidence_for_watchlist(candidates)[:5]
|
||||
|
||||
fallback = [
|
||||
item
|
||||
for item in evidence
|
||||
if item.get("id")
|
||||
and item.get("operatorStatus") not in NON_CONTRIBUTING_OPERATOR_STATUSES
|
||||
]
|
||||
return _sorted_evidence_for_watchlist(fallback)[:5]
|
||||
|
||||
|
||||
def _sorted_evidence_for_watchlist(evidence: list[dict[str, Any]]) -> list[dict[str, Any]]:
|
||||
return sorted(
|
||||
evidence,
|
||||
key=lambda item: (
|
||||
1 if item.get("contributed", True) else 0,
|
||||
float(item.get("confidence", 0) or 0),
|
||||
str(item.get("retrievedAt", "")),
|
||||
),
|
||||
reverse=True,
|
||||
)
|
||||
|
||||
|
||||
def _knowledge_type_value(value: str) -> str:
|
||||
normalized = value.strip() or "other"
|
||||
aliases = {
|
||||
"public_figure": "celebrity",
|
||||
"rejected_reference": "rejected_image",
|
||||
}
|
||||
return aliases.get(normalized, normalized)
|
||||
|
||||
|
||||
def _knowledge_entry_type(value: str) -> KnowledgeEntryType:
|
||||
try:
|
||||
return KnowledgeEntryType(_knowledge_type_value(value))
|
||||
except ValueError:
|
||||
return KnowledgeEntryType.OTHER
|
||||
|
||||
|
||||
|
|
|
|||
Loading…
Reference in a new issue