爬虫学习笔记整理一

tips

• 不论爬取哪个网页，都可以加上请求头信息

requests使用代理

import requests

url = "http://httpbin.org/ip"#访问这个地址会返回访问者的ip地址
proxies = {'http':'119.39.68.252:8118'}
resp = requests.get(url, proxies=proxies)

print(resp.text)

lxml是用C语言写的，没有提示

1. 解析html字符串，使用etree.HTML(htmlstr)进行解析
2. 解析html文件，使用etree.parse(filepath,parser=etree.HTMLParser(encoding="utf-8"))

• print(resp.text)#返回的是经过解码后的字符串，是str（unicode）类型，有时候会出现解码为乱码的情况，这时就需要自己指定解码方式
• print(resp.content)#返回的是一个原生的字符串，就是从网页上抓取的没有经过处理的字符串，是bytes类型

resp = requests.get(url, headers=HEADERS)
text = resp.content.decode("gbk", errors="ignore")  #加上errors='ignore'，不然会报错

xpath语法

trs = table.xpath(".//tr[position()>2]")
for i in range(1,5):
    url = "https://www.gushiwen.org/default_%s.aspx" % i
    parse_page(url)
    
 imgs = html.xpath("//a[@class='col-xs-6 col-sm-3']//img[@class!='gif']")    
    for img in imgs:
        img_url = img.get('data-original')#获取标签的某个属性值

json字符串操作

在python中，只有基本数据类型才可以转换为json格式的字符串，即
int、float、str、list、dict、tuple

import json

# 1.将python对象转换成json串
persons = [
    {'name':'张三', 'age':18, 'gender':'男'},
    {'name':'张四', 'age':17, 'gender':'女'},
    {'name':'张五', 'age':19, 'gender':'男'}
    ]

json_str = json.dumps(persons)
print(type(json_str), json_str)

# 2.将json数据存储到文件
with open("person.json", "w", encoding="utf-8") as f:
    #f.write(json_str)#方式一
    json.dump(persons, f, ensure_ascii=False)#方式二：直接将python对象dump到文件中，就不用再转换一步为字符串


# 将json字符串load成python对象
# 3.直接从文件中读取
jsonstr = '[{"name": "u5f20u4e09", "age": 18, "gender": "u7537"}, {"name": "u5f20u56db", "age": 17, "gender": "u5973"}, {"name": "u5f20u4e94", "age": 19, "gender": "u7537"}]'
persons = json.loads(jsonstr)
for p in persons:
    print(p)
  
print("-" * 50)

# 5.直接从文件读取
with open("person.json", "r", encoding="utf-8") as f:
    persons = json.load(f)
    print(type(persons))
    for p in persons:
        print(p)

存储为csv文件

import csv

# =============================================================================
# 读取csv文件
# =============================================================================
def read_csv_by_index():
    """以索引模式读取"""
    with open("5-冀东水泥集团问题汇总【36家】.csv", "r", encoding="utf-8") as f:
        #reader其实是一个迭代器
        reader = csv.reader(f)
        next(reader)#指针向下挪动一行
        for row in reader:
            hazard = {}
            # print(row)
            express = row[1]
            ytype = row[2]
            company = row[-1]
            if express != '' and ytype != '' and company != '':
                hazard = {'express': express, 'ytype': ytype, 'company': company}
                print(hazard)

def read_csv_by_dict():
    with open("5-冀东水泥集团问题汇总【36家】.csv", "r", encoding="utf-8") as f:
        #使用DicReader创建的不会包含第一行表头，此时的迭代器reader返回的是一个字典
        reader = csv.DictReader(f)
        for line in reader:
            print(line)
    pass
    
# =============================================================================
# 写入数据到csv文件
# =============================================================================
def write_list_to_csv():
    #定义表头
    headers = ['username','age','sex']
    
    #要写入的数据
    data = [
        ('张三', 12, '男'),    
        ('李三', 19, '男'),
        ('张五', 28, '男'),
        ('王小二', 18, '女'),
    ]
    
    #encoding="utf-8", newline=""两个参数分别解决乱码和多余空行的问题
    with open("person_list.csv", "w", encoding="utf-8", newline="") as f:
        writer = csv.writer(f)
        writer.writerow(headers)#写入表头
        writer.writerows(data)#写入数据
    pass

def write_dict_to_csv():
    headers = ['username','age','sex']
    data = [
        {'username':'张三', 'age': 18, 'sex':'男'},    
        {'username':'李三', 'age': 16, 'sex':'女'},    
        {'username':'张五', 'age': 18, 'sex':'女'},    
        {'username':'王小二', 'age': 19, 'sex':'男'},    
    ]
    
    with open("person_dict.csv", "w", encoding="utf-8", newline="") as f:
        writer = csv.DictWriter(f, headers)
        writer.writeheader() #这种字典方式需要调用方法才能写入表头       
        writer.writerows(data)
    pass


if __name__ == "__main__":
    write_dict_to_csv()

正则模块

title = re.sub(r'[?？.。！!]', '', title)#处理掉文件路径中不符合要求的字符
suffix = os.path.splitext(img_url)[1]#获取图片后缀名

#写入图片到文件
data = requests.get(img_url).content
    with open("images/"+filename, "wb") as f:
        f.write(data)

selenium+chromedriver

chromedriver下载地址：
http://chromedriver.storage.googleapis.com/index.html
根据谷歌浏览器版本下载对应的驱动chromedriver

import time
from selenium import webdriver

def automation(url):
    # driver_path是谷歌浏览器驱动chromedriver.exe的存放路径
    driver_path = "F:\python\chromedriver_win32\chromedriver.exe"
    driver = webdriver.Chrome(executable_path=driver_path)
    
    #接下来就可以使用driver去操作谷歌浏览器了
    driver.get(url)
    
    #获取网页源代码(一般都是通过属性)
    print(driver.page_source)
    """关闭浏览器"""
    time.sleep(5)
    # driver.close()#退出当前页面
    driver.quit()#退出浏览器

    """获取元素的方法一"""
    #获取输入框
    inputTag = driver.find_element_by_id("kw")
    inputTag.send_keys("python")


if __name__ == "__main__":
    url = "http://www.baidu.com"
    automation(url)