From 30857d391d3a5343f9f98d0d8f317662eeef2c79 Mon Sep 17 00:00:00 2001 From: jxxghp Date: Tue, 15 Sep 2026 21:53:13 +0800 Subject: [PATCH] =?UTF-8?q?refactor(application):=20=E6=94=B6=E6=95=9B?= =?UTF-8?q?=E5=85=B1=E4=BA=AB=E9=87=8D=E5=A4=8D=E9=80=BB=E8=BE=91?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit --- app/application/classification/enrichment.py | 24 +------- app/application/classification/migration.py | 20 +------ app/application/classification/projection.py | 18 +----- app/application/download/classification.py | 11 +--- app/application/download/organization.py | 4 +- app/application/download/tasks.py | 9 +-- app/application/download/validation.py | 13 ++++ app/application/server/payload.py | 24 ++++++++ app/application/server/report.py | 16 +---- app/application/server/share.py | 16 +---- app/domain/classification/conditions.py | 30 ++++++++++ tests/test_application_shared_utilities.py | 63 ++++++++++++++++++++ tests/test_architecture_dependencies.py | 1 + 13 files changed, 148 insertions(+), 101 deletions(-) create mode 100644 app/application/download/validation.py create mode 100644 app/application/server/payload.py create mode 100644 app/domain/classification/conditions.py create mode 100644 tests/test_application_shared_utilities.py diff --git a/app/application/classification/enrichment.py b/app/application/classification/enrichment.py index cda42f35f..c021694d2 100644 --- a/app/application/classification/enrichment.py +++ b/app/application/classification/enrichment.py @@ -16,6 +16,7 @@ from typing import Any, Protocol, TypeAlias, cast from pydantic import ValidationError from app.domain.classification.evaluator import read_fact +from app.domain.classification.conditions import condition_field_ids from app.domain.classification.fields import ( classification_fact_matches_definition, field_definition_map, @@ -429,7 +430,7 @@ def _referenced_missing_standard_fields( continue if rule.sources and media_source not in rule.sources: continue - for field_id in _condition_field_ids(rule.when): + for field_id in condition_field_ids(rule.when): if field_id.startswith(("identity.", "extensions.")): continue definition = definitions.get(field_id) @@ -441,27 +442,6 @@ def _referenced_missing_standard_fields( return tuple(fields) -def _condition_field_ids(node: ClassificationConditionNode) -> tuple[str, ...]: - """按策略声明顺序递归返回条件树叶子字段。""" - if isinstance(node, ClassificationCondition): - return (node.field,) - if not isinstance(node, ClassificationConditionGroup): - return () - if node.all is not None: - children: Sequence[ClassificationConditionNode] = node.all - elif node.any is not None: - children = node.any - elif node.not_ is not None: - children = (node.not_,) - else: - children = () - return tuple( - field_id - for child in children - for field_id in _condition_field_ids(child) - ) - - def _provider_fields( provider: ClassificationEnrichmentProvider, missing_fields: Sequence[str], diff --git a/app/application/classification/migration.py b/app/application/classification/migration.py index f21ef43e2..005fc680e 100644 --- a/app/application/classification/migration.py +++ b/app/application/classification/migration.py @@ -22,6 +22,7 @@ from app.domain.classification.vocabulary import ( from app.domain.classification.vocabulary import ( TMDB_GENRE_KEYS as _TMDB_GENRE_KEYS, ) +from app.domain.classification.conditions import condition_field_ids from app.schemas.category import ( CategoryConfig, CategoryRule, @@ -222,7 +223,7 @@ def legacy_extension_fields_from_policy( """按策略条件和别名重建可直接注册的 TMDB 旧比较扩展字段声明。""" context = _MigrationContext() for rule in policy.rules: - for field_id in _condition_field_ids(rule.when): + for field_id in condition_field_ids(rule.when): if not field_id.startswith(_EXTENSION_PREFIX): continue field_name = field_id.removeprefix(_EXTENSION_PREFIX) @@ -392,7 +393,7 @@ def _migrate_media_categories( if any( field_id.startswith(_EXTENSION_PREFIX) for node in nodes - for field_id in _condition_field_ids(node) + for field_id in condition_field_ids(node) ) else [] ) @@ -943,21 +944,6 @@ def _error_count(diagnostics: Sequence[LegacyClassificationDiagnostic]) -> int: return sum(item.severity == "error" for item in diagnostics) -def _condition_field_ids(node: ClassificationConditionNode) -> list[str]: - """按条件树顺序提取全部叶子字段 ID。""" - if isinstance(node, ClassificationCondition): - return [node.field] - if node.all is not None: - children = node.all - elif node.any is not None: - children = node.any - elif node.not_ is not None: - children = [node.not_] - else: - children = [] - return [field_id for child in children for field_id in _condition_field_ids(child)] - - def _append_unique(values: list[str], value: str) -> None: """按首次出现顺序追加非重复字符串。""" if value not in values: diff --git a/app/application/classification/projection.py b/app/application/classification/projection.py index 7b07f318c..1829facef 100644 --- a/app/application/classification/projection.py +++ b/app/application/classification/projection.py @@ -16,6 +16,7 @@ from app.application.classification.migration import ( LegacyDiagnosticPathPart, LegacyMediaKey, ) +from app.domain.classification.conditions import condition_field_ids from app.domain.classification.vocabulary import TMDB_GENRE_KEYS from app.schemas.category import ( CategoryConfig, @@ -473,7 +474,7 @@ def _controlled_tmdb_fields( field_ids: list[str] = [] if isinstance(policy_or_field_defs, ClassificationPolicy): for rule in policy_or_field_defs.rules: - field_ids.extend(_condition_field_ids(rule.when)) + field_ids.extend(condition_field_ids(rule.when)) else: for definition in policy_or_field_defs: if definition.source_support.get(_TMDB_SOURCE) == "extension": @@ -489,21 +490,6 @@ def _controlled_tmdb_fields( return tuple(fields) -def _condition_field_ids(node: ClassificationConditionNode) -> list[str]: - """按条件树顺序提取全部叶子字段 ID。""" - if isinstance(node, ClassificationCondition): - return [node.field] - if node.all is not None: - children = node.all - elif node.any is not None: - children = node.any - elif node.not_ is not None: - children = [node.not_] - else: - children = [] - return [field_id for child in children for field_id in _condition_field_ids(child)] - - def _project_legacy_tmdb_field( field_name: str, tmdb_info: Mapping[str, object], diff --git a/app/application/download/classification.py b/app/application/download/classification.py index 6bb2f4250..074761ab1 100644 --- a/app/application/download/classification.py +++ b/app/application/download/classification.py @@ -13,6 +13,7 @@ from app.application.directory import ( build_media_download_path, validate_download_save_path, ) +from app.application.download.validation import validate_torrent_hash from app.application.history import DownloadHistorySnapshot from app.domain.classification.validation import validate_classification_category_path from app.domain.context import MediaInfo, MusicInfo @@ -161,14 +162,6 @@ class DownloadSourceClassificationService: self._update_torrent = update_torrent self._resolve_plan = resolve_plan - @staticmethod - def _validate_hash(hash_value: str) -> None: - """校验 BitTorrent v1 Hash，拒绝模糊任务定位。""" - if len(hash_value) != 40 or any( - character not in "0123456789abcdefABCDEF" for character in hash_value - ): - raise ValueError("hash 格式无效") - def plan( self, *, @@ -178,7 +171,7 @@ class DownloadSourceClassificationService: media_category: Optional[str] = None, ) -> dict[str, Any]: """生成分类计划，只有 execute 为真时才请求下载器移动。""" - self._validate_hash(hash_value) + validate_torrent_hash(hash_value) torrents = self._list_torrents( hashs=[hash_value], downloader=downloader, diff --git a/app/application/download/organization.py b/app/application/download/organization.py index b7d01d65b..1efb735b0 100644 --- a/app/application/download/organization.py +++ b/app/application/download/organization.py @@ -6,6 +6,7 @@ from typing import Any from app.application.configuration import get_configured_system_config from app.application.directory import DirectoryHelper, validate_download_save_path +from app.application.download.validation import validate_torrent_hash from app.domain.classification.validation import validate_classification_category_path from app.domain.meta.metabase import MetaBase from app.domain.meta.metamusic import MetaMusic @@ -286,8 +287,7 @@ def _rename_qb_root(chain: Any, downloader: str, hash_value: str, old_name: str, def organize_existing_source(hash_value: str, request: Any, chain: Any, media_chain: Any) -> dict[str, Any]: """生成可重放的预览计划，确认后仅通过下载器修改任务路径。""" - if not re.fullmatch(r"[0-9a-fA-F]{40}", hash_value): - raise ValueError("hash 格式无效") + validate_torrent_hash(hash_value) history = chain.download_history_repository.get_by_hash(hash_value) if history is None: raise ValueError("未找到该任务的下载历史") diff --git a/app/application/download/tasks.py b/app/application/download/tasks.py index daa113b79..79fbfc692 100644 --- a/app/application/download/tasks.py +++ b/app/application/download/tasks.py @@ -3,6 +3,7 @@ from typing import Any, Callable, List, Optional from app.application.directory import validate_download_save_path +from app.application.download.validation import validate_torrent_hash from app.application.history import DownloadHistorySnapshot from app.schemas.transfer import DownloaderTorrent, DownloadTaskMedia from app.schemas.types import TorrentStatus @@ -99,12 +100,6 @@ class DownloadTaskMutationService: self._set_downloading = set_downloading self._update_torrent = update_torrent - @staticmethod - def _validate_hash(hash_value: str) -> None: - """校验 BitTorrent v1 Hash，拒绝模糊任务定位。""" - if len(hash_value) != 40 or any(character not in "0123456789abcdefABCDEF" for character in hash_value): - raise ValueError("hash 格式无效") - @staticmethod def _normalize_list(values: Optional[list[str]]) -> Optional[list[str]]: """移除字符串列表中的空项。""" @@ -137,7 +132,7 @@ class DownloadTaskMutationService: seeding_time_limit: Optional[int] = None, ) -> dict[str, Any]: """校验并执行启停、标签、限速、Tracker 与保存位置修改。""" - self._validate_hash(hash_value) + validate_torrent_hash(hash_value) if action not in {None, "start", "stop"}: raise ValueError("action 只支持 start 或 stop") tags = self._normalize_list(tags) diff --git a/app/application/download/validation.py b/app/application/download/validation.py new file mode 100644 index 000000000..6ffd468fb --- /dev/null +++ b/app/application/download/validation.py @@ -0,0 +1,13 @@ +"""下载应用服务共享的输入校验。""" + +from __future__ import annotations + +import re + +_TORRENT_HASH_PATTERN = re.compile(r"^[0-9a-fA-F]{40}$") + + +def validate_torrent_hash(hash_value: str) -> None: + """校验 BitTorrent v1 Hash，拒绝模糊任务定位。""" + if not _TORRENT_HASH_PATTERN.fullmatch(hash_value): + raise ValueError("hash 格式无效") diff --git a/app/application/server/payload.py b/app/application/server/payload.py new file mode 100644 index 000000000..11b18414f --- /dev/null +++ b/app/application/server/payload.py @@ -0,0 +1,24 @@ +"""中心服务订阅载荷的共享投影。""" + +from __future__ import annotations + +from collections.abc import Collection +from typing import Any + +from app.schemas.media import resolve_media_identity + + +def build_subscribe_payload( + item: object, + fields: Collection[str], +) -> dict[str, Any] | None: + """过滤本地订阅字段并规范化中心服务使用的媒体身份。""" + if not isinstance(item, dict): + return None + media_source, media_id = resolve_media_identity(media=item) + if not media_source or not media_id: + return None + payload = {key: value for key, value in item.items() if key in fields} + payload["media_source"] = str(media_source) + payload["media_id"] = media_id + return payload diff --git a/app/application/server/report.py b/app/application/server/report.py index 6cf88af90..f4a241546 100644 --- a/app/application/server/report.py +++ b/app/application/server/report.py @@ -5,7 +5,7 @@ from __future__ import annotations from collections.abc import Awaitable, Callable from typing import Any, Optional -from app.schemas.media import resolve_media_identity +from app.application.server.payload import build_subscribe_payload class ServerReportService: @@ -78,19 +78,7 @@ class ServerReportService: def build_subscribe_payload(self, item: Optional[dict]) -> Optional[dict]: """构造中心服务订阅统计载荷并移除本地运行字段。""" - if not isinstance(item, dict): - return None - media_source, media_id = resolve_media_identity(media=item) - if not media_source or not media_id: - return None - payload = { - key: value - for key, value in item.items() - if key in self.SUBSCRIBE_FIELDS - } - payload["media_source"] = str(media_source) - payload["media_id"] = media_id - return payload + return build_subscribe_payload(item, self.SUBSCRIBE_FIELDS) def build_plugin_payload( self, diff --git a/app/application/server/share.py b/app/application/server/share.py index d7edd75fe..6329c3d9f 100644 --- a/app/application/server/share.py +++ b/app/application/server/share.py @@ -8,7 +8,7 @@ from dataclasses import asdict, dataclass from typing import Any, Optional from app.application.workflow import WorkflowSnapshot -from app.schemas.media import resolve_media_identity +from app.application.server.payload import build_subscribe_payload @dataclass(frozen=True, slots=True) @@ -67,19 +67,7 @@ class ServerSharingService: def build_subscribe_payload(self, item: Optional[dict]) -> Optional[dict]: """构造订阅分享载荷并隔离本地字段和旧专用 ID。""" - if not isinstance(item, dict): - return None - media_source, media_id = resolve_media_identity(media=item) - if not media_source or not media_id: - return None - payload = { - key: value - for key, value in item.items() - if key in self.SUBSCRIBE_FIELDS - } - payload["media_source"] = str(media_source) - payload["media_id"] = media_id - return payload + return build_subscribe_payload(item, self.SUBSCRIBE_FIELDS) @staticmethod def prepare_workflow(workflow: WorkflowSnapshot) -> dict: diff --git a/app/domain/classification/conditions.py b/app/domain/classification/conditions.py new file mode 100644 index 000000000..5db5e9169 --- /dev/null +++ b/app/domain/classification/conditions.py @@ -0,0 +1,30 @@ +"""分类条件树的共享纯函数。""" + +from __future__ import annotations + +from app.schemas.category import ( + ClassificationCondition, + ClassificationConditionGroup, + ClassificationConditionNode, +) + + +def condition_field_ids(node: ClassificationConditionNode) -> tuple[str, ...]: + """按条件树声明顺序递归返回全部叶子字段 ID。""" + if isinstance(node, ClassificationCondition): + return (node.field,) + if not isinstance(node, ClassificationConditionGroup): + return () + if node.all is not None: + children = node.all + elif node.any is not None: + children = node.any + elif node.not_ is not None: + children = (node.not_,) + else: + children = () + return tuple( + field_id + for child in children + for field_id in condition_field_ids(child) + ) diff --git a/tests/test_application_shared_utilities.py b/tests/test_application_shared_utilities.py new file mode 100644 index 000000000..68fc22bd4 --- /dev/null +++ b/tests/test_application_shared_utilities.py @@ -0,0 +1,63 @@ +"""Application 重复逻辑收敛后的共享纯函数回归测试。""" + +from types import SimpleNamespace + +import pytest + +from app.application.download.validation import validate_torrent_hash +from app.application.server.payload import build_subscribe_payload +from app.domain.classification.conditions import condition_field_ids +from app.schemas.category import ClassificationCondition, ClassificationConditionGroup + + +def test_condition_field_ids_preserves_nested_declaration_order() -> None: + """条件树公共遍历必须保留 all/any/not 的声明顺序。""" + node = ClassificationConditionGroup( + all=[ + ClassificationCondition(field="media.title", operator="exists"), + ClassificationConditionGroup( + any=[ + ClassificationCondition(field="media.year", operator="exists"), + ClassificationConditionGroup( + not_=ClassificationCondition( + field="media.adult", + operator="exists", + ) + ), + ] + ), + ] + ) + + assert condition_field_ids(node) == ( + "media.title", + "media.year", + "media.adult", + ) + + +@pytest.mark.parametrize("value", ["", "a" * 39, "g" * 40, "a" * 41]) +def test_validate_torrent_hash_rejects_non_v1_hash(value: str) -> None: + """下载应用服务必须拒绝长度或字符集不符合 v1 Hash 的输入。""" + with pytest.raises(ValueError, match="hash 格式无效"): + validate_torrent_hash(value) + + +def test_build_subscribe_payload_filters_fields_and_requires_identity() -> None: + """中心服务公共载荷投影只保留白名单字段并要求媒体身份完整。""" + payload = build_subscribe_payload( + { + "name": "Demo", + "media_source": "tmdb", + "media_id": "123", + "private": "hidden", + }, + {"name", "media_source", "media_id"}, + ) + + assert payload == { + "name": "Demo", + "media_source": "themoviedb", + "media_id": "123", + } + assert build_subscribe_payload(SimpleNamespace(), {"name"}) is None diff --git a/tests/test_architecture_dependencies.py b/tests/test_architecture_dependencies.py index dcf5f2c59..3e89a6e7d 100644 --- a/tests/test_architecture_dependencies.py +++ b/tests/test_architecture_dependencies.py @@ -392,6 +392,7 @@ def test_domain_classification_is_a_pure_direct_import_package() -> None: package = APP_ROOT / "domain" / "classification" assert {path.name for path in package.glob("*.py")} == { "__init__.py", + "conditions.py", "evaluator.py", "facts.py", "fields.py",