feat(ingest): db-only library ingest and telegram manual ingest

tests/test_library.py

@@ -0,0 +1,296 @@
+# -*- coding: utf-8 -*-
+"""入库（候选确认 / 手动）、忽略、拉黑动作测试（假只读客户端，不联网）:
+uv run python tests/test_library.py
+"""
+
+import json
+import os
+import sys
+import tempfile
+from pathlib import Path
+
+sys.path.insert(0, os.path.dirname(os.path.dirname(os.path.abspath(__file__))))
+sys.stdout.reconfigure(encoding="utf-8")
+
+from fixtures import view_for  # noqa: E402
+from reaction_board.bilibili.client import BilibiliError  # noqa: E402
+from reaction_board.bilibili.models import VideoDetail  # noqa: E402
+from reaction_board.config import ConfigStore  # noqa: E402
+from reaction_board.database import Database  # noqa: E402
+from reaction_board.repository import Repository  # noqa: E402
+from reaction_board.services.library import LibraryService  # noqa: E402
+from reaction_board.timeutil import shanghai_date  # noqa: E402
+
+NOW = 1_793_444_393
+DATE = shanghai_date(NOW)
+BV = "BV1xvxi6LEZD"
+
+
+def detail(
+    aid,
+    *,
+    bvid=BV,
+    title="咻咻满 reaction",
+    up_mid=100,
+    up_name="UP100",
+    pubdate=NOW - 3600,
+    view=819,
+    duration=9504,
+    pages=2,
+):
+    payload = view_for(aid, view=view)
+    data = payload["data"]
+    data["bvid"] = bvid
+    data["title"] = title
+    data["pubdate"] = pubdate
+    data["ctime"] = pubdate
+    data["duration"] = duration
+    data["owner"] = {"mid": up_mid, "name": up_name, "face": ""}
+    data["pages"] = [{"cid": 1, "page": i + 1} for i in range(pages)]
+    return VideoDetail.from_api(data)
+
+
+class FakeReadClient:
+    def __init__(self, by_bvid=None, by_aid=None, short_map=None, fail_bvids=()):
+        self.by_bvid = by_bvid or {}
+        self.by_aid = by_aid or {}
+        self.short_map = short_map or {}
+        self.fail_bvids = set(fail_bvids)
+
+    def view_detail(self, *, aid=None, bvid=None):
+        if bvid is not None:
+            if bvid in self.fail_bvids:
+                raise BilibiliError(f"view 失败: {bvid}")
+            if bvid in self.by_bvid:
+                return self.by_bvid[bvid]
+            raise BilibiliError(f"未知 bvid: {bvid}")
+        if aid in self.by_aid:
+            return self.by_aid[aid]
+        raise BilibiliError(f"未知 aid: {aid}")
+
+    def resolve_short_link(self, url):
+        if url not in self.short_map:
+            raise BilibiliError(f"未知短链: {url}")
+        return self.short_map[url]
+
+
+class FakeProvider:
+    def __init__(self, client):
+        self.client = client
+
+    def get(self):
+        return self.client
+
+
+class FakeUpProfile:
+    def __init__(self):
+        self.calls = []
+
+    def capture(self, mids, *, names=None, live_since=None, backfilled=False):
+        self.calls.append((list(mids), dict(names or {})))
+
+        class Result:
+            requested = len(mids)
+            updated_videos = len(mids)
+            failed = 0
+
+        return Result()
+
+
+def write_config(tmp):
+    data = {
+        "bilibili": {
+            "read": {"mid": 1, "csrf": "r", "cookie": "read-cookie"},
+            "write": {"mid": 2, "csrf": "w", "cookie": "write-cookie"},
+        },
+        "storage": {
+            "database_path": str(Path(tmp) / "app.db"),
+            "backup_dir": str(Path(tmp) / "backup"),
+        },
+        "discover": {"keywords": ["x"], "pages": 1, "window_days": 14},
+    }
+    path = Path(tmp) / "config.json"
+    path.write_text(json.dumps(data, ensure_ascii=False), encoding="utf-8")
+    return path
+
+
+def setup(tmp, client, up_profile=None):
+    store = ConfigStore(write_config(tmp), env={})
+    db = Database(Path(tmp) / "app.db", backup_dir=Path(tmp) / "backup")
+    db.initialize()
+    repo = Repository(db, clock=lambda: NOW)
+    service = LibraryService(
+        store, repo, FakeProvider(client),
+        up_profile_service=up_profile, clock=lambda: NOW,
+    )
+    return db, repo, service
+
+
+def add_candidate(repo, aid, *, up_mid=100, pubdate=NOW - 3600):
+    return repo.upsert_candidate(
+        aid=aid, bvid=BV, title="咻咻满 reaction", up_mid=up_mid,
+        up_name=f"UP{up_mid}", pubdate=pubdate, cover=None, score=3,
+        channel="search:test#p1", now=NOW,
+    )
+
+
+def test_ingest_candidate_added_and_snapshot():
+    with tempfile.TemporaryDirectory() as tmp:
+        ups = FakeUpProfile()
+        client = FakeReadClient(by_aid={9: detail(9)})
+        db, repo, service = setup(tmp, client, ups)
+        try:
+            add_candidate(repo, 9)
+            result = service.ingest_candidate(9)
+            assert result.ok is True and result.state == "added"
+            assert result.aid == 9
+            assert repo.get_candidate(9) is None
+            video = repo.get_video(9)
+            assert video["status"] == "active"
+            assert video["source"] == "candidate:telegram"
+            assert video["folder_id"] is None
+            assert video["up_mid"] == 100
+            snap = repo.get_snapshot(DATE, 9)
+            assert snap is not None and snap["ok"] == 1 and snap["view"] == 819
+            assert ups.calls == [([100], {100: "UP100"})]
+        finally:
+            db.close()
+
+
+def test_ingest_candidate_exists_and_missing():
+    with tempfile.TemporaryDirectory() as tmp:
+        client = FakeReadClient(by_aid={9: detail(9)})
+        db, repo, service = setup(tmp, client)
+        try:
+            add_candidate(repo, 9)
+            assert service.ingest_candidate(9).state == "added"
+            # 候选已被删除，但视频已在库
+            again = service.ingest_candidate(9)
+            assert again.ok is True and again.state == "exists"
+            missing = service.ingest_candidate(999)
+            assert missing.ok is False and missing.state == "missing"
+        finally:
+            db.close()
+
+
+def test_ingest_candidate_recovers_dropped():
+    with tempfile.TemporaryDirectory() as tmp:
+        client = FakeReadClient(by_aid={9: detail(9)})
+        db, repo, service = setup(tmp, client)
+        try:
+            add_candidate(repo, 9)
+            assert service.ingest_candidate(9).state == "added"
+            repo.set_video_status(9, "dropped", now=NOW)
+            add_candidate(repo, 9)
+            result = service.ingest_candidate(9)
+            assert result.ok is True and result.state == "recovered"
+            assert repo.get_video(9)["status"] == "active"
+        finally:
+            db.close()
+
+
+def test_ingest_candidate_falls_back_when_detail_fails():
+    with tempfile.TemporaryDirectory() as tmp:
+        client = FakeReadClient(fail_bvids={BV})
+        db, repo, service = setup(tmp, client)
+        try:
+            add_candidate(repo, 9)
+            result = service.ingest_candidate(9)
+            assert result.ok is True and result.state == "added"
+            assert repo.get_video(9)["title"] == "咻咻满 reaction"
+            assert repo.get_snapshot(DATE, 9) is None
+        finally:
+            db.close()
+
+
+def test_manual_ingest_forms_and_multiple():
+    with tempfile.TemporaryDirectory() as tmp:
+        ups = FakeUpProfile()
+        client = FakeReadClient(
+            by_bvid={BV: detail(9)},
+            by_aid={12345: detail(77, bvid="BV77", up_mid=200, up_name="UP200")},
+            short_map={"https://b23.tv/abc": f"https://www.bilibili.com/video/{BV}"},
+        )
+        db, repo, service = setup(tmp, client, ups)
+        try:
+            result = service.ingest_text(
+                f"看这个 https://www.bilibili.com/video/{BV} 还有 av12345"
+            )
+            assert result.refs == 2 and result.added == 2
+            assert [r.aid for r in result.results] == [9, 77]
+            assert repo.get_video(9)["source"] == "manual:telegram"
+            assert repo.get_video(77)["status"] == "active"
+
+            short = service.ingest_text("https://b23.tv/abc")
+            assert short.added == 0
+            assert short.results[0].state == "exists"
+
+            none = service.ingest_text("没有视频")
+            assert none.refs == 0 and none.results == []
+        finally:
+            db.close()
+
+
+def test_manual_ingest_error_isolation():
+    with tempfile.TemporaryDirectory() as tmp:
+        client = FakeReadClient(
+            by_bvid={BV: detail(9)},
+            fail_bvids={"BVfail000000"},
+        )
+        db, repo, service = setup(tmp, client)
+        try:
+            result = service.ingest_text(f"{BV} BVfail000000")
+            assert result.added == 1
+            states = [r.state for r in result.results]
+            assert states == ["added", "error"]
+        finally:
+            db.close()
+
+
+def test_snapshot_not_overwritten():
+    with tempfile.TemporaryDirectory() as tmp:
+        client = FakeReadClient(by_bvid={BV: detail(9, view=819)})
+        db, repo, service = setup(tmp, client)
+        try:
+            repo.upsert_snapshot(
+                DATE, 9, metrics={"view": 5000}, ok=True, err=None, captured_at=NOW,
+            )
+            service.ingest_text(BV)
+            snap = repo.get_snapshot(DATE, 9)
+            assert snap["view"] == 5000  # 已有成功快照不覆盖
+        finally:
+            db.close()
+
+
+def test_ignore_and_blacklist():
+    with tempfile.TemporaryDirectory() as tmp:
+        client = FakeReadClient()
+        db, repo, service = setup(tmp, client)
+        try:
+            add_candidate(repo, 11, up_mid=300)
+            add_candidate(repo, 12, up_mid=300)
+            assert service.ignore(11).state == "rejected"
+            assert repo.get_candidate(11)["decision"] == "rejected"
+            result = service.blacklist(12)
+            assert result.ok is True
+            assert repo.is_blacklisted(300)
+            assert repo.get_candidate(12)["decision"] == "rejected"
+            assert service.ignore(999).state == "missing"
+        finally:
+            db.close()
+
+
+def main():
+    test_ingest_candidate_added_and_snapshot()
+    test_ingest_candidate_exists_and_missing()
+    test_ingest_candidate_recovers_dropped()
+    test_ingest_candidate_falls_back_when_detail_fails()
+    test_manual_ingest_forms_and_multiple()
+    test_manual_ingest_error_isolation()
+    test_snapshot_not_overwritten()
+    test_ignore_and_blacklist()
+    print("OK: 入库、手动入库、忽略、拉黑测试全部通过")
+
+
+if __name__ == "__main__":
+    main()