python3的UnicodeDecodeError

爬虫部分解码异常

response.content.decode() # 默认使用 utf-8 出现解码异常

以下是设计的通用解码

通过 text 获取编码

# 通过 text 获取编码
import requests
from lxml import etree


def public_decode():
    headers = {
        'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/78.0.3904.108 Safari/537.36'
    }
    response = requests.get('https://blog.csdn.net/a13951206104', headers=headers)
    html = etree.HTML(response.text) # response.text 能自动获取编码, 大多乱码
    _charset = html.xpath('//@charset') or []
    if _charset:
        encode_content = response.content.decode(_charset[0].strip().lower(),
                                                 errors='replace')  # 如果设置为replace，则会用?取代非法字符；
        return {'response_text': encode_content, "response_obj": response}
    for _charset_ in ['utf-8', 'gbk', 'gb2312'] # 国内主要这3种:
        if '�' not in response.content.decode(_charset_, errors='replace'):
            return {'response_text': response.content.decode(_charset_, errors='replace'),
                    "response_obj": response}
        else:
            # 默认还得是 utf-8
            return {'response_text': response.content.decode('utf-8', errors='replace'),
                    "response_obj": response}
	    

通过数据 来解编码(推荐)

def public_decode(response):
    headers = {
        'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/78.0.3904.108 Safari/537.36'
    }
    response = requests.get('https://blog.csdn.net/a13951206104', headers=headers)
    html = etree.HTML(response.text)
    # 不希望抓下来的数据中有非法字符
    item = dict()
    result = None
    for _charset_ in ['utf-8', 'gbk', 'gb2312']:
        if response:
            result = response.content.decode(_charset_, errors='replace')
            item['content'] =  html.xpath('//*[@id="content"]')
            if '�' not in result['content'].strip():
                result =response.content.decode(_charset_, errors='replace')
                break
    if not result:
        # 默认 utf-8
        result = response.content.decode(_charset_, errors='replace')

errors=‘replace’ 使解码不报异常, 然后把几个常用的编码一个个试下, 最后要看落下来的数据, 所以最好拿数据 去获取合适的编码