python大佬请进，关于抓取的数据格式的问题

张大牛

最近百度上线的chat  https://chat.baidu.com/

想自己套个壳玩玩，原打算是用PHP的，但是因为这个结果是流数据，PHP的curl貌似搞不定，只好转到python

代码已完成90%

baidu.rar (1.24 KB, 下载次数: 8)

16 小时前 上传

点击文件名下载附件

代码运行成功，也有返回数据

但返回的数据都是这种格式

C:\>python baidu.py
请求成功
event:ping

event:message
data:{"status":0,"qid":"12643291455431891488","pkgId":"cd971053-9663-4d88-8a54-2
c428ddbf3b7_0","sessionId":"43d95fc2-b94f-442f-9c5b-b8078f712862","isDefault":1,
"isShow":0,"data":{"message":{"msgId":"cd971053-9663-4d88-8a54-2c428ddbf3b7","is
Rebuild":false,"updateTime":"1694966929746","metaData":{"state":"waiting-resp","
endTurn":false,"userInfo":{"status":3}},"content":{}}}}


event:message
data:{"status":0,"qid":"12643291455431891488","pkgId":"cd971053-9663-4d88-8a54-2
c428ddbf3b7_1","sessionId":"43d95fc2-b94f-442f-9c5b-b8078f712862","isDefault":1,
"isShow":0,"data":{"message":{"msgId":"cd971053-9663-4d88-8a54-2c428ddbf3b7","is
Rebuild":false,"updateTime":"1694966933900","metaData":{"state":"waiting-resp","
endTurn":false,"userInfo":{"status":3}},"content":{"searchQuery":{"querys":["鲁
迅是谁"]}}}}}


event:message
data:{"status":0,"qid":"12643291455431891488","pkgId":"cd971053-9663-4d88-8a54-2
c428ddbf3b7_2","sessionId":"43d95fc2-b94f-442f-9c5b-b8078f712862","isDefault":1,
"isShow":0,"data":{"message":{"msgId":"cd971053-9663-4d88-8a54-2c428ddbf3b7","is
Rebuild":false,"updateTime":"1694966933978","metaData":{"state":"generating-resp
","endTurn":false,"userInfo":{"status":3}},"content":{"generator":{"text":"鲁迅
，原名周樟","type":"txt","showType":"append","antiFlag":0,"isFinished":false}}}}
}


event:message
data:{"status":0,"qid":"12643291455431891488","pkgId":"cd971053-9663-4d88-8a54-2
c428ddbf3b7_3","sessionId":"43d95fc2-b94f-442f-9c5b-b8078f712862","isDefault":1,
"isShow":0,"data":{"message":{"msgId":"cd971053-9663-4d88-8a54-2c428ddbf3b7","is
Rebuild":false,"updateTime":"1694966934570","metaData":{"state":"generating-resp
","endTurn":false,"userInfo":{"status":3}},"content":{"generator":{"text":"寿，
后改名周树人，字豫山，后改字豫才，是浙江绍兴的人。","type":"txt","showType":"app
end","antiFlag":0,"isFinished":false}}}}}

请问怎么才能把需要的text内容提取出来，组成完整的答案？

求大佬指教

taiyi747

正则万能，你这个格式我前两天刚处理过，把内容交给ai让他给你写正则表达式就好了

BackDoor

python 有  json库直接输出的。

parsed_data['data']['message']['content']['generator']['text']

张大牛

taiyi747 发表于 2023-9-18 00:21
正则万能，你这个格式我前两天刚处理过，把内容交给ai让他给你写正则表达式就好了 ...

谢谢大佬，没想到呢，一直想着json解析

张大牛

BackDoor 发表于 2023-9-18 00:24
python 有  json库直接输出的。

parsed_data['data']['message']['content']['generator']['text']

好的，谢谢大佬

william2ct

json.loads()

zqm840527

这不就是JSON？

秋秋0827

这是py的json基础问题呀，你的头像有点花里胡哨呀