How to Extract Text from JSON‑Like Strings in Python Using Regex and split()

Introduction

Hello, I'm a Python enthusiast. A fan asked how to extract text from a JSON‑like string obtained during web crawling.

Idea

Regular expressions are a quick way to pull out the desired fields.

Implementation

1. Using regex

# -*- coding: utf-8 -*-
import re
text = """
Top2 26
Top2 "word":"经纪人不得为假唱假演奏提供条件"
Top2 "query":"经纪人不得为假唱假演奏提供条件"
Top2 "show":[]
Top2 "desc":"18日，文旅部发布关于《演出经纪人员管理办法(征求意见稿)》公开征求意见的公告。"
Top2 "img":"https://fyb-1.cdn.bcebos.com/fyb-1//5b4bc1de60744e69f34225af1452a395"
Top2 "url":"https://www.baidu.com/s?wd=..."
Top2 "rawUrl":"https://www.baidu.com/s?wd=..."
Top2 "hotScore":"2325661"
Top2 "hotChange":"same"
Top2 "hotTag":"0"
Top2 "appUrl":"https://www.baidu.com/s?wd=..."
"""
regex = re.findall(r'":"(.*?)"', text)
for data in regex:
    print(data)

Running the script prints the extracted values.

2. Using split()

# -*- coding: utf-8 -*-
import re
text = """
Top2 26
Top2 "word":"经纪人不得为假唱假演奏提供条件"
Top2 "query":"经纪人不得为假唱假演奏提供条件"
Top2 "show":[]
Top2 "desc":"18日，文旅部发布关于《演出经纪人员管理办法(征求意见稿)》公开征求意见的公告。"
Top2 "img":"https://fyb-1.cdn.bcebos.com/fyb-1//5b4bc1de60744e69f34225af1452a395"
Top2 "url":"https://www.baidu.com/s?wd=..."
Top2 "rawUrl":"https://www.baidu.com/s?wd=..."
Top2 "hotScore":"2325661"
Top2 "hotChange":"same"
Top2 "hotTag":"0"
Top2 "appUrl":"https://www.baidu.com/s?wd=..."
"""
raw_text = text.strip('
').split('Top2 ')
for line in raw_text:
    print(line.split('":"')[-1].replace('"', ''))

This method works but is less flexible and more brittle.

Conclusion

The article demonstrates how to process a raw string from a Python web crawler using either regular expressions or the split() function, satisfying the fan's request. Readers are encouraged to experiment with other techniques and share their results.