feat: 提交热榜评论分析工具 MVP 基线

2026-07-03 11:12:12 +08:00
parent 0231be6338
commit 9bad7b6e63
60 changed files with 4058 additions and 0 deletions
@@ -0,0 +1,84 @@
+from typing import Any
+
+from app.platforms.base import CommentData, ContentItemData, HotspotData, TikHubClient, first_present, parse_timestamp
+
+
+class XiaohongshuPlatform:
+    def __init__(self, client: TikHubClient | None) -> None:
+        self.client = client
+
+    def map_hotspots(self, payload: dict[str, Any], *, limit: int) -> list[HotspotData]:
+        items = payload.get("data", {}).get("data", {}).get("items", [])
+        hotspots = []
+        for index, item in enumerate(items[:limit], start=1):
+            hotspots.append(
+                HotspotData(
+                    source_hot_id=str(item.get("id")) if item.get("id") is not None else None,
+                    title=str(item.get("title") or ""),
+                    rank=index,
+                    heat_value=str(item.get("score")) if item.get("score") is not None else None,
+                    raw_data=item,
+                )
+            )
+        return hotspots
+
+    def map_items(self, payload: dict[str, Any], *, limit: int) -> list[ContentItemData]:
+        raw_items = payload.get("data", {}).get("data", {}).get("items", [])
+        notes = [item.get("note", item) for item in raw_items]
+        notes.sort(key=lambda note: 0 if int(note.get("comments_count") or 0) > 0 else 1)
+        result = []
+        for note in notes[:limit]:
+            note_id = note.get("id")
+            if not note_id:
+                continue
+            result.append(
+                ContentItemData(
+                    source_item_id=str(note_id),
+                    item_type="note",
+                    title=note.get("title"),
+                    summary=note.get("desc"),
+                    url=note.get("url"),
+                    raw_data=note,
+                )
+            )
+        return result
+
+    def map_comments(self, payload: dict[str, Any], *, limit: int) -> list[CommentData]:
+        comments = payload.get("data", {}).get("data", {}).get("comments", [])
+        result = []
+        for comment in comments[:limit]:
+            content = first_present(comment, "content", "text")
+            if not content:
+                continue
+            result.append(
+                CommentData(
+                    source_comment_id=first_present(comment, "comment_id", "id"),
+                    content=str(content),
+                    author=self._author_name(comment),
+                    like_count=comment.get("like_count"),
+                    comment_time=parse_timestamp(first_present(comment, "create_time", "create_time_str")),
+                    raw_data=comment,
+                )
+            )
+        return result
+
+    def fetch_hotspots(self, *, limit: int) -> list[HotspotData]:
+        return self.map_hotspots(self.client.get("/api/v1/xiaohongshu/web_v2/fetch_hot_list"), limit=limit)
+
+    def search_items_by_hotspot(self, keyword: str, *, limit: int) -> list[ContentItemData]:
+        payload = self.client.get(
+            "/api/v1/xiaohongshu/app_v2/search_notes",
+            params={"keyword": keyword, "page": 1, "sort": "general", "note_type": 0},
+        )
+        return self.map_items(payload, limit=limit)
+
+    def fetch_comments(self, source_item_id: str, *, limit: int) -> list[CommentData]:
+        payload = self.client.get(
+            "/api/v1/xiaohongshu/app_v2/get_note_comments",
+            params={"note_id": source_item_id, "cursor": "", "index": 0, "pageArea": "UNFOLDED", "sort_strategy": "latest_v2"},
+        )
+        return self.map_comments(payload, limit=limit)
+
+    def _author_name(self, comment: dict[str, Any]) -> str | None:
+        user = comment.get("user_info") or comment.get("user") or {}
+        return user.get("nickname") or user.get("name") if isinstance(user, dict) else None