feat: 小红书支持通过博主ID采集笔记和评论,小红书type=search时支持配置按哪种排序方式获取笔记数据,小红书笔记增加视频地址和标签字段

This commit is contained in:
jayeeliu@gmail.com
2024-03-02 01:49:42 +08:00
parent c09f9fef68
commit 61ba8c5cc7
8 changed files with 244 additions and 2 deletions

View File

@@ -31,12 +31,20 @@ async def update_xhs_note(note_item: Dict):
user_info = note_item.get("user", {})
interact_info = note_item.get("interact_info", {})
image_list: List[Dict] = note_item.get("image_list", [])
tag_list: List[Dict] = note_item.get("tag_list", [])
video_url = ''
if note_item.get('type') == 'video':
videos = note_item.get('video').get('media').get('stream').get('h264')
if type(videos).__name__ == 'list':
video_url = ','.join([ v.get('master_url') for v in videos])
local_db_item = {
"note_id": note_item.get("note_id"),
"type": note_item.get("type"),
"title": note_item.get("title") or note_item.get("desc", "")[:255],
"desc": note_item.get("desc", ""),
"video_url": video_url,
"time": note_item.get("time"),
"last_update_time": note_item.get("last_update_time", 0),
"user_id": user_info.get("user_id"),
@@ -48,6 +56,7 @@ async def update_xhs_note(note_item: Dict):
"share_count": interact_info.get("share_count"),
"ip_location": note_item.get("ip_location", ""),
"image_list": ','.join([img.get('url', '') for img in image_list]),
"tag_list": ','.join([tag.get('name', '') for tag in tag_list if tag.get('type')=='topic']),
"last_modify_ts": utils.get_current_timestamp(),
"note_url": f"https://www.xiaohongshu.com/explore/{note_id}"
}
@@ -77,3 +86,32 @@ async def update_xhs_note_comment(note_id: str, comment_item: Dict):
}
utils.logger.info(f"[store.xhs.update_xhs_note_comment] xhs note comment:{local_db_item}")
await XhsStoreFactory.create_store().store_comment(local_db_item)
async def save_creator(user_id: str, creator: Dict):
user_info = creator.get('basicInfo', {})
follows = 0
fans = 0
interaction = 0
for i in creator.get('interactions'):
if i.get('type') == 'follows':
follows = i.get('count')
elif i.get('type') == 'fans':
fans = i.get('count')
elif i.get('type') == 'interaction':
interaction = i.get('count')
local_db_item = {
'user_id': user_id,
'nickname': user_info.get('nickname'),
'gender': '' if user_info.get('gender') == 1 else '' ,
'avatar': user_info.get('images'),
'desc': user_info.get('desc'),
'ip_location': user_info.get('ip_location'),
'follows': follows,
'fans': fans,
'interaction': interaction,
'tag_list': json.dumps({tag.get('tagType'):tag.get('name') for tag in creator.get('tags')}),
}
utils.logger.info(f"[store.xhs.save_creator] creator:{local_db_item}")
await XhsStoreFactory.create_store().store_creator(local_db_item)

View File

@@ -25,6 +25,7 @@ class XHSNote(XhsBaseModel):
type = fields.CharField(null=True, max_length=16, description="笔记类型(normal | video)")
title = fields.CharField(null=True, max_length=255, description="笔记标题")
desc = fields.TextField(null=True, description="笔记描述")
video_url = fields.TextField(null=True, description="视频地址")
time = fields.BigIntField(description="笔记发布时间戳", index=True)
last_update_time = fields.BigIntField(description="笔记最后更新时间戳")
liked_count = fields.CharField(null=True, max_length=16, description="笔记点赞数")
@@ -32,6 +33,7 @@ class XHSNote(XhsBaseModel):
comment_count = fields.CharField(null=True, max_length=16, description="笔记评论数")
share_count = fields.CharField(null=True, max_length=16, description="笔记分享数")
image_list = fields.TextField(null=True, description="笔记封面图片列表")
tag_list = fields.TextField(null=True, description="标签列表")
note_url = fields.CharField(null=True, max_length=255, description="笔记详情页的URL")
class Meta:
@@ -55,3 +57,19 @@ class XHSNoteComment(XhsBaseModel):
def __str__(self):
return f"{self.comment_id} - {self.content}"
class XhsCreator(XhsBaseModel):
desc = fields.TextField(null=True, description="用户描述")
gender = fields.CharField(null=True, max_length=1, description="性别")
follows = fields.CharField(null=True, max_length=16, description="关注数")
fans = fields.CharField(null=True, max_length=16, description="粉丝数")
interaction = fields.CharField(null=True, max_length=16, description="获赞和收藏数")
# follows = fields.IntField(description="关注数")
# fans = fields.IntField(description="粉丝数")
# interaction = fields.IntField(description="获赞和收藏数")
tag_list = fields.TextField(null=True, description="标签列表") # json字符串
class Meta:
table = "xhs_creator"
table_description = "小红书博主"

View File

@@ -72,6 +72,17 @@ class XhsCsvStoreImplement(AbstractStore):
"""
await self.save_data_to_csv(save_item=comment_item, store_type="comments")
async def store_creator(self, creator: Dict):
"""
Xiaohongshu content CSV storage implementation
Args:
creator: creator dict
Returns:
"""
await self.save_data_to_csv(save_item=creator, store_type="creator")
class XhsDbStoreImplement(AbstractStore):
async def store_content(self, content_item: Dict):
@@ -121,6 +132,31 @@ class XhsDbStoreImplement(AbstractStore):
comment_pydantic.model_validate(comment_data)
await XHSNoteComment.filter(comment_id=comment_id).update(**comment_data.model_dump())
async def store_creator(self, creator: Dict):
"""
Xiaohongshu content DB storage implementation
Args:
creator: creator dict
Returns:
"""
from .xhs_store_db_types import XhsCreator
user_id = creator.get("user_id")
if not await XhsCreator.filter(user_id=user_id).first():
creator["add_ts"] = utils.get_current_timestamp()
creator["last_modify_ts"] = creator["add_ts"]
creator_pydantic = pydantic_model_creator(XhsCreator, name="CreatorPydanticCreate", exclude=('id',))
creator_data = creator_pydantic(**creator)
creator_pydantic.model_validate(creator_data)
await XhsCreator.create(**creator_data.model_dump())
else:
creator["last_modify_ts"] = utils.get_current_timestamp()
creator_pydantic = pydantic_model_creator(XhsCreator, name="CreatorPydanticUpdate", exclude=('id', 'add_ts',))
creator_data = creator_pydantic(**creator)
creator_pydantic.model_validate(creator_data)
await XhsCreator.filter(user_id=user_id).update(**creator_data.model_dump())
class XhsJsonStoreImplement(AbstractStore):
json_store_path: str = "data/xhs"
@@ -181,3 +217,14 @@ class XhsJsonStoreImplement(AbstractStore):
"""
await self.save_data_to_json(comment_item, "comments")
async def store_creator(self, creator: Dict):
"""
Xiaohongshu content JSON storage implementation
Args:
creator: creator dict
Returns:
"""
await self.save_data_to_json(creator, "creator")