1

server/app/services/collection_service.py

@@ -14,6 +14,7 @@ logger = logging.getLogger(__name__)
 
 # 内容源配置（域名轮换时只需改这里）
 SOURCE_KEY = "mdcmai4"
+SOURCE_CODE = "madou"     # sources 表中的 code，用于给导入内容打「来源」标
 UPSTREAM = "https://mdcmai4.xyz"
 PRODUCT_ID = 1
 CATEGORY_ID = 12          # 本站分类：AI短剧
@@ -37,6 +38,21 @@ def _proxy_url(path: str) -> str:
     return f"{SELF_API_BASE}/api/v1/m3u8/proxy?path={quote(path, safe='')}"
 
 
+async def default_source_id(db: AsyncSession) -> int:
+    """麻豆等内容源的来源 id（sources.code 匹配；缺失时取第一个启用来源）"""
+    from app.models.video import Source
+
+    sid = (
+        await db.execute(select(Source.id).where(Source.code == SOURCE_CODE))
+    ).scalar_one_or_none()
+    if sid:
+        return sid
+    sid = (
+        await db.execute(select(Source.id).where(Source.status == 1).order_by(Source.sort, Source.id))
+    ).scalars().first()
+    return sid or 0
+
+
 async def _import_drama(client: httpx.AsyncClient, db: AsyncSession, drama_id: int) -> tuple[int, int]:
     """导入单部短剧，返回 (新增集数, 跳过集数)"""
     detail = await _fetch_json(client, f"{UPSTREAM}/api/v1/short-dramas/{drama_id}", {"productId": PRODUCT_ID})
@@ -63,6 +79,7 @@ async def _import_drama(client: httpx.AsyncClient, db: AsyncSession, drama_id: i
             description=(detail.get("description") or "")[:500],
             author_id=AUTHOR_USER_ID,
             category_id=CATEGORY_ID,
+            source_id=await default_source_id(db),
             total_episodes=0,
             view_count=int(detail.get("heatCount") or 0),
             status=1,
@@ -94,6 +111,7 @@ async def _import_drama(client: httpx.AsyncClient, db: AsyncSession, drama_id: i
             Video(
                 user_id=AUTHOR_USER_ID,
                 category_id=CATEGORY_ID,
+                source_id=series.source_id,
                 series_id=series.id,
                 episode_no=ep_no,
                 title=(ep.get("titleOverride") or ep.get("title") or f"第{ep_no}集")[:100],
@@ -209,6 +227,7 @@ async def _import_video_item(db: AsyncSession, source: CollectionSource, item: d
     v = Video(
         user_id=AUTHOR_USER_ID,
         category_id=source.category_id,
+        source_id=source.source_id or await default_source_id(db),
         title=(item.get("title") or f"视频{vid}")[:100],
         description=(item.get("description") or "")[:500],
         cover_url=_abs_cover(item)[:500],
@@ -232,13 +251,23 @@ async def sync_source(
     max_pages: int = 0,
     stop_after_existing: int = 30,
     delay: float = 0.25,
+    should_cancel=None,
 ) -> dict:
-    """按源的 URL 模板翻页采集；连续 stop_after_existing 条已存在则提前停止（增量）"""
-    stat = {"created": 0, "skipped": 0, "pages": 0, "stopped_early": False}
+    """按源的 URL 模板翻页采集；连续 stop_after_existing 条已存在则提前停止（增量）。
+    should_cancel 返回 True 时每页/每条边界退出（用于手动取消）"""
+    stat = {"created": 0, "skipped": 0, "pages": 0, "stopped_early": False, "cancelled": False}
     consecutive_existing = 0
+
+    def _cancelled() -> bool:
+        return bool(should_cancel and should_cancel())
+
     async with httpx.AsyncClient(follow_redirects=True) as client:
         page = 1
         while True:
+            if _cancelled():
+                stat["cancelled"] = True
+                logger.info("通用采集源 %s 同步被取消（第 %s 页前）", source.id, page)
+                return stat
             url = source.list_url.replace("{{page}}", str(page))
             data = await _fetch_json(client, url)
             items = data.get("items", [])
@@ -246,6 +275,10 @@ async def sync_source(
                 break
             stat["pages"] = page
             for item in items:
+                if _cancelled():
+                    stat["cancelled"] = True
+                    logger.info("通用采集源 %s 同步被取消（处理 item 中）", source.id)
+                    return stat
                 if await _import_video_item(db, source, item):
                     stat["created"] += 1
                     consecutive_existing = 0