用 selenium 爬 bjidex.com 商品数据，无论试多少遍都只能拿到 40 条。脑壳扣了半天，想不明白。

V2EX = way to explore

V2EX 是一个关于分享和探索的地方

已注册用户请登录

我的代码：

from selenium import webdriver
from selenium.webdriver.common.by import By
from selenium.webdriver.support.ui import WebDriverWait
from selenium.webdriver.support import expected_conditions as EC
import pandas as pd
import time

# 设置 Chrome 选项以启用 headless 模式和自定义 user-agent
chrome_options = webdriver.ChromeOptions()
chrome_options.add_argument("--headless")
chrome_options.add_argument(
    f'--user-agent="Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/128.0.0.0 Safari/537.36"'
)

# 初始化 WebDriver
driver = webdriver.Chrome(
    executable_path="D:/lab/chromedriver-win64/chromedriver.exe", options=chrome_options
)  
url = "https://webs.bjidex.com/sys-bsc-home/#/bscConsole/tradingMarket"

# 访问网页
driver.get(url)

# 初始化一个列表来保存数据
data_list = []

# 爬取数据
for page in range(184):

    # 构建翻页按钮的 XPath
    if page < 4:
        next_button_xpath = "/html/body/div[2]/div[2]/div/div/div[2]/div/div/section/div/div/div/div[7]/div/div[2]/ul/li[10]/button"
    elif 4 < page < 181:
        next_button_xpath = "/html/body/div[2]/div[2]/div/div/div[2]/div/div/section/div/div/div/div[7]/div/div[2]/ul/li[12]/button"
    elif 181 < page:
        next_button_xpath = "/html/body/div[2]/div[2]/div/div/div[2]/div/div/section/div/div/div/div[7]/div/div[2]/ul/li[10]/button"
    else:
        next_button_xpath = "/html/body/div[2]/div[2]/div/div/div[2]/div/div/section/div/div/div/div[7]/div/div[2]/ul/li[11]/button"

    # 爬取每页的 10 组数据
    for i in range(1, 11):
        time.sleep(1)  # 等待页面加载新内容
        # 构建每组的 XPath
        product_name_xpath = f"/html/body/div[2]/div[2]/div/div/div[2]/div/div/section/div/div/div/div[7]/div/div[1]/div/ul/li[{i}]/div[1]/div/div[1]/div/span[1]"
        supplier_list_xpath = f"/html/body/div[2]/div[2]/div/div/div[2]/div/div/section/div/div/div/div[7]/div/div[1]/div/ul/li[{i}]/div[1]/div/div[3]/span"
        product_type_xpath = f"/html/body/div[2]/div[2]/div/div/div[2]/div/div/section/div/div/div/div[7]/div/div[1]/div/ul/li[{i}]/div[1]/div/div[1]/div/span[2]"
        application_scenario_xpath = f"/html/body/div[2]/div[2]/div/div/div[2]/div/div/section/div/div/div/div[7]/div/div[1]/div/ul/li[{i}]/div[1]/div/div[3]/div"
        product_description_xpath = f"/html/body/div[2]/div[2]/div/div/div[2]/div/div/section/div/div/div/div[7]/div/div[1]/div/ul/li[{i}]/div[1]/div/div[2]"
        price_xpath = f"/html/body/div[2]/div[2]/div/div/div[2]/div/div/section/div/div/div/div[7]/div/div[1]/div/ul/li[{i}]/div[2]/div[1]"

        try:
            # 供应商提供商品名称
            product_name = driver.find_element(By.XPATH, product_name_xpath).text

            # 数据供应商名单
            supplier_list = driver.find_element(By.XPATH, supplier_list_xpath).text

            # 商品类型
            product_type = driver.find_element(By.XPATH, product_type_xpath).text

            # 应用场景
            application_scenario = driver.find_element(
                By.XPATH, application_scenario_xpath
            ).text

            # 商品描述
            product_description = driver.find_element(
                By.XPATH, product_description_xpath
            ).text

            # 价格
            price = driver.find_element(By.XPATH, price_xpath).text

            # 将数据添加到列表
            data_list.append(
                {
                    "页数": page + 1,
                    "供应商提供商品名称": product_name,
                    "数据供应商名单": supplier_list,
                    "商品类型": product_type,
                    "应用场景": application_scenario,
                    "商品描述": product_description,
                    "价格": price,
                }
            )
        except Exception as e:
            print(f"Error on page {page + 1}, item {i}: {e}")

    # 点击翻页按钮
    try:
        next_button = WebDriverWait(driver, 2).until(  # 等待时间为 2 秒
            EC.element_to_be_clickable((By.XPATH, next_button_xpath))
        )
        next_button.click()
    except Exception as e:
        print("翻页出错或已经是最后一页:", e)
        break  # 如果无法翻页，则跳出循环

# 关闭浏览器
driver.quit()

# 将列表转换为 DataFrame
data_df = pd.DataFrame(data_list)

# 输出为表格
data_df.to_csv(
    "bjidex.com_data.csv", index=False, encoding="utf_8_sig"
)  # 保存为 CSV 文件
print(data_df)  # 打印 DataFrame

输出

PS D:\lab\bigdata24.9.9> & C:/tools/miniconda3/python.exe d:/lab/bigdata24.9.9/bjidex.com.py

DevTools listening on ws://127.0.0.1:60057/devtools/browser/66a61aa5-3598-4069-94bd-d4f10be20d96
[42892:8184:0915/232611.717:ERROR:ssl_client_socket_impl.cc(882)] handshake failed; returned -1, SSL error code 1, net_error -101
[42892:8184:0915/232611.834:ERROR:ssl_client_socket_impl.cc(882)] handshake failed; returned -1, SSL error code 1, net_error -101
[42892:8184:0915/232630.213:ERROR:ssl_client_socket_impl.cc(882)] handshake failed; returned -1, SSL error code 1, net_error -101
翻页出错或已经是最后一页: Message: 

    页数  ...       价格
0    1  ...   0.5 元/次
1    1  ...     0 元/次
2    1  ...     0 元/次
3    1  ...   2.5 元/次
...
37   4  ...   0.2 元/次
38   4  ...   0.1 元/次
39   4  ...  0.15 元/次

[40 rows x 7 columns]

8 条回复 • 2024-09-17 00:02:31 +08:00

mumbler

3 天前

用桌面去调试，去掉 headless ，看看页面是否显示正常

xe2vdw

3 天前 via Android

elif 4 < page < 181:
没有＝ 4

HUZHUANGZHUANG

2 天前

找个 AI 给你看看代码问题

c137Toma

2 天前

@xe2vdw else 对应 = 4 和 = 181

c137Toma

2 天前

我看到第 4 页和第 181 页的时候翻页按钮会变，for 循环里面是不是 page = 3 对应第 4 页，page = 180 对应第 181 页？

1423

2 天前

这是工作? 还是学生?
工作的话, 这也太菜了

blackeeper

2 天前

直接调接口不香吗？
``` python
import requests

url = 'https://webs.bjidex.com/api/dstp/data-asset-server/dataProduct/deal/list'
data = {"searchName":"","productType":"","industry":"","viewCode":1,"pageSize":10,"pageNum":8}
headers = {
'User-Agent' : 'Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36',
'Referer' : 'https://webs.bjidex.com/sys-bsc-home/',
'Origin' : 'https://webs.bjidex.com'
}
response = requests.post(url, json=data,headers=headers)
print(response.text)
```

c137Toma

2 天前

是学生，谢谢各位。