时间戳

首页 > 解决方案 > Chromedriver 在抓取时不断更改时区

python - Chromedriver 在抓取时不断更改时区

问题描述

下面是我的 Python 代码的开头，它成功地从该网站上抓取了所有表格信息并将其导出到 CSV 文件。我在这个刮刀上遇到的唯一问题是 Chromedriver 不断更改右上角的时区，这最终通过分配一些日期不正确的游戏来扭曲我的输出。我尝试在页面源中查找允许我单击“GMT-8 太平洋时区”的链接或标签，但不幸的是我找不到任何东西。令人沮丧的是，当我将 URL 复制并粘贴到浏览器中时，Chrome 会立即切换回太平洋时区。有谁知道在使用 Chromedriver 抓取数据时如何解决这个时区问题？提前致谢！

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 re
import pandas as pd

# set scope and create empty lists
year = 2018
lastpage = 50
Date = []
Time = []
Team1 = []
Team2 = []
Score = []
All_ML = []
Team1_ML = []
Team2_ML = []

driver = webdriver.Chrome()
driver.get('http://www.oddsportal.com/')
driver.execute_script('op.selectTimeZone(6);')

# set up for loop to loop through all pages
for x in range(1, lastpage + 1):
    url = "http://www.oddsportal.com/baseball/usa/mlb-" + str(year) + "/results/#/page/" + str(x) + "/'"
    driver.get(url)


    # wait until java table loads and then grab data
    element = WebDriverWait(driver, 10).until(
        EC.visibility_of_element_located((By.XPATH, '//*[@id="tournamentTable"]')))
    odds = element.text
    print (odds)

    # close temporary chrome screen
    driver.close()

    # reformat resulting text for consistency
    odds = re.sub("[0-9] - ", str(year)[-1] + " -- ", odds)
    odds = re.sub(" - ", "\nteam2", odds)

    # split text by line
    odds = odds.split("\n")

    counter = 1

    # set up loop to classify each line of text
    for line in odds:

        # if a game was abandoned or cancelled, set score to N/A
        if re.match(".*( {1})[a-zA-Z]*\.$", line):
            Score.append("N/A")

            # if date format is matched, add to date list and reset counter
        if re.match("(.{2} .{3} .{4}.*)", line):
            currdate = line[:11]
            Date.append(currdate)
            counter = 1

        # if time format is matched at beginning of string, add time to list, add team1 to list, check if there was a new date for this game. if not, add current date from previous game
        elif re.match('(.{2}:.{2})', line):
            Time.append(line[:5])
            Team1.append(line[6:])
            if counter > 1:
                Date.append(currdate)
            counter += 1

        # if its a team2 line, add to team2 list. if score is on the same line, add to score list
        elif re.match("team2.*", line):
            if re.match(".*:.*", line):
                Score.append(re.sub("[a-zA-Z]* *", "", line[-5:]))
                Team2.append(re.sub(" {1}[0-9]*:[0-9]*", "", line[5:len(line)]))
            else:
                Team2.append(re.sub(" {1}[a-zA-Z]*\.", "", line[5:]))

        # if score is on it's own line, add to score list
        elif re.match(".*:.*", line):
            Score.append(re.sub(" ", "", line))

        # add all moneylines to a list
        elif re.match("[+\-.*]", line):
            All_ML.append(line)

    # add odd money lines to list1, even moneylines to list 2
    Team1_ML = All_ML[0::2]
    Team2_ML = All_ML[1::2]

# create dataframe with all lists
df = pd.DataFrame(
    {'Date': Date,
     'Time': Time,
     'Team1': Team1,
     'Team2': Team2,
     'Score': Score,
     'Team1_ML': Team1_ML,
     'Team2_ML': Team2_ML})

# save
df.to_csv('odds2018.csv')

标签： pythonweb-scrapingselenium-chromedriver

解决方案

为了充实 pguardiario 的注释，如果您使用 Chrome devtools 查看右上角的按钮，每个按钮都会触发指向某个时区代码https://www.oddsportal.com/set-timezone/n/所在位置的链接。n这些函数实际上触发了一个函数op.selectTimeZone(n)，它会改变你在屏幕上的时区。您可以在您的 Chrome 控制台中输入op.selectTimeZone(n).

如果这对您有用，您可以通过使用模拟控制台 javascript 调用来合并它，n所选时区的代码在哪里：

driver.execute_script('op.selectTimeZone(n);')

您可以在每次驱动程序初始化调用后添加它，以强制设置时区，例如：

for x in range(1, lastpage + 1):
    url = "http://www.oddsportal.com/baseball/usa/mlb-" + str(year) + "/results/#/page/" + str(x) + "/'"

   driver = webdriver.Chrome()
    driver.get(url)

    # Set timezone
    driver.execute_script('op.selectTimeZone(6);')

    # wait until java table loads and then grab data
    element = WebDriverWait(driver, 10).until(
        EC.visibility_of_element_located((By.XPATH, '//*[@id="tournamentTable"]')))
    odds = element.text

请注意，您可能需要设置等待计时器，因为您在选择时区之后添加了额外的执行。

此外，您确实不需要为每个循环重置驱动程序调用，除非您计划并行化for循环。如果您将驱动程序初始化并关闭循环，这可能会运行得更快。

编辑：

因此，如果您直接访问结果页面，您将无法在不触发页面重新加载的情况下设置时区。您可能需要将设置和加载排除在循环之外，例如

driver = webdriver.Chrome()
driver.get('http://www.oddsportal.com/')
# Proc JS on-click for timezone selection button
driver.execute_script("op.showHideTimeZone();ElementSelect.expand( 'user-header-timezone' , 'user-header-timezone-expander' , null , function(){op.hideTimeZone()} );this.blur();")
driver.execute_script('op.selectTimeZone(6);')

for x in range(1, lastpage + 1):
    url = "http://www.oddsportal.com/baseball/usa/mlb-" + str(year) + "/results/#/page/" + str(x) + "/'"

    driver.get(url)

    # wait until java table loads and then grab data
    element = WebDriverWait(driver, 10).until(
        EC.visibility_of_element_located((By.XPATH, '//*[@id="tournamentTable"]')))
    odds = element.text
    print(odds)
# close temporary chrome screen
driver.close()

推荐阅读