refactor: 数据存储重构，分离不同类型的存储实现

2026-06-06 09:57:25 +08:00 · 2024-01-14 22:06:31 +08:00
parent e31aebbdfb
commit 894dabcf63
37 changed files with 1427 additions and 864 deletions
--- a/store/weibo/init.py
+++ b/store/weibo/init.py
@@ -0,0 +1,88 @@
+# -*- coding: utf-8 -*-
+# @Author  : relakkes@gmail.com
+# @Time    : 2024/1/14 21:34
+# @Desc    :
+
+from typing import List
+
+import config
+
+from .weibo_store_db_types import *
+from .weibo_store_impl import *
+
+
+class WeibostoreFactory:
+    STORES = {
+        "csv": WeiboCsvStoreImplement,
+        "db": WeiboDbStoreImplement,
+        "json": BiliJsonStoreImplement
+    }
+
+    @staticmethod
+    def create_store() -> AbstractStore:
+        store_class = WeibostoreFactory.STORES.get(config.SAVE_DATA_OPTION)
+        if not store_class:
+            raise ValueError(
+                "[WeibotoreFactory.create_store] Invalid save option only supported csv or db or json ...")
+        return store_class()
+
+async def update_weibo_note(note_item: Dict):
+    mblog: Dict = note_item.get("mblog")
+    user_info: Dict = mblog.get("user")
+    note_id = mblog.get("id")
+    save_content_item = {
+        # 微博信息
+        "note_id": note_id,
+        "content": mblog.get("text"),
+        "create_time": utils.rfc2822_to_timestamp(mblog.get("created_at")),
+        "create_date_time": str(utils.rfc2822_to_china_datetime(mblog.get("created_at"))),
+        "liked_count": str(mblog.get("attitudes_count", 0)),
+        "comments_count": str(mblog.get("comments_count", 0)),
+        "shared_count": str(mblog.get("reposts_count", 0)),
+        "last_modify_ts": utils.get_current_timestamp(),
+        "note_url": f"https://m.weibo.cn/detail/{note_id}",
+        "ip_location": mblog.get("region_name", "").replace("发布于 ", ""),
+
+        # 用户信息
+        "user_id": str(user_info.get("id")),
+        "nickname": user_info.get("screen_name", ""),
+        "gender": user_info.get("gender", ""),
+        "profile_url": user_info.get("profile_url", ""),
+        "avatar": user_info.get("profile_image_url", ""),
+    }
+    utils.logger.info(
+        f"[store.weibo.update_weibo_note] weibo note id:{note_id}, title:{save_content_item.get('content')[:24]} ...")
+    await WeibostoreFactory.create_store().store_content(content_item=save_content_item)
+
+
+async def batch_update_weibo_note_comments(note_id: str, comments: List[Dict]):
+    if not comments:
+        return
+    for comment_item in comments:
+        await update_weibo_note_comment(note_id, comment_item)
+
+
+async def update_weibo_note_comment(note_id: str, comment_item: Dict):
+    comment_id = str(comment_item.get("id"))
+    user_info: Dict = comment_item.get("user")
+    save_comment_item = {
+        "comment_id": comment_id,
+        "create_time": utils.rfc2822_to_timestamp(comment_item.get("created_at")),
+        "create_date_time": str(utils.rfc2822_to_china_datetime(comment_item.get("created_at"))),
+        "note_id": note_id,
+        "content": comment_item.get("text"),
+        "sub_comment_count": str(comment_item.get("total_number", 0)),
+        "comment_like_count": str(comment_item.get("like_count", 0)),
+        "last_modify_ts": utils.get_current_timestamp(),
+        "ip_location": comment_item.get("source", "").replace("来自", ""),
+
+        # 用户信息
+        "user_id": str(user_info.get("id")),
+        "nickname": user_info.get("screen_name", ""),
+        "gender": user_info.get("gender", ""),
+        "profile_url": user_info.get("profile_url", ""),
+        "avatar": user_info.get("profile_image_url", ""),
+    }
+    utils.logger.info(
+        f"[store.weibo.update_weibo_note_comment] Weibo note comment: {comment_id}, content: {save_comment_item.get('content','')[:24]} ...")
+    await WeibostoreFactory.create_store().store_comment(comment_item=save_comment_item)