我怎样才能让 selenium 与 Fivethirtyeight 搜索栏进行交互？答案

【问题标题】：How can I get selenium to interact with fivethirtyeight search bars?我怎样才能让 selenium 与 Fivethirtyeight 搜索栏进行交互？
【发布时间】：2022-01-20 15:24:01
【问题描述】：

我正在使用这个网站 (https://projects.fivethirtyeight.com/2020-nba-player-projections/) 并尝试自动搜索玩家。我有一个玩家列表，但似乎无法连接到正确的元素。我尝试过 xpath、id 和 css_selector。下面是一个示例代码：

driver.get('https://projects.fivethirtyeight.com/2020-nba-player-projections/')
search_bar = driver.find_element_by_id('player-selectbox')
search_bar.click() 
search_bar.send_keys('lebron james')
search_bar.send_keys(Keys.RETURN)

无论我尝试哪种方式，我都会收到以下错误：

Message: element not interactable

下面是我正在尝试访问的框的屏幕截图的链接。

screenshot

谢谢！

【问题讨论】：

您能告诉我们您正在尝试与哪个元素进行交互吗？添加带有搜索栏指示器的屏幕截图会很有用

标签： python selenium search

【解决方案1】：

from selenium import webdriver
from selenium.webdriver.common.keys import Keys
import time


driver.get("https://projects.fivethirtyeight.com/2020-nba-player- projections/")
    
driver.find_element_by_xpath("//input[@placeholder='player name']").click()
    
driver.find_element_by_xpath("//input[@placeholder='player name']").send_keys('Kyrie Irving')

driver.find_element_by_xpath("//input[@placeholder='player name']").send_keys(Keys.ARROW_RIGHT)
time.sleep(10)
driver.close()

这段代码完美运行，复杂之处在于它是一个自动填充类型的搜索栏，我找到了一个简单的解决方法，使用 Keys.ARROW_RIGHT 来处理它

下面附上截图，

【讨论】：

【解决方案2】：

这应该可以工作

search_bar = driver.find_element_by_xpath("//input[@placeholder='player name']")
search_bar.click()
search_bar.send_keys('lebron james')
time.sleep(2)
driver.find_element_by_xpath("//*[@class='tt-suggestion tt-selectable']").click()

请随意将time.sleep 替换为explicit wait

【讨论】：

谢谢@Anand 这在大多数情况下都有效，我会随机捕获少数例外情况
我试过了，还是不行

【解决方案3】：

为什么不直接从 json 中获取数据呢？会比 Selenium 高效得多。

import requests
import re
import json
import pandas as pd
from fuzzywuzzy import process

# Get Player Slugs from javascript
response = requests.get('https://projects.fivethirtyeight.com/2020-nba-player-projections/js/bundle.js')

js = response.text
p = re.compile('{t\.exports=(\[.*}]},{}\])')
result = p.search(js).group(1).rsplit('},{}',1)[0]

result = result.replace('team_abbr', '"team_abbr"')
result = result.replace('name', '"name"')
result = result.replace('full_team', '"full_team"')
result = result.replace('short_team', '"short_team"')
result = result.replace('slug', '"slug"')
result = result.replace(',randable:!1', '')
result = result.replace(',randable:!0', '')

playerSearch = json.loads(result)
df = pd.DataFrame(playerSearch)

players = list(df['name'])
slugs = list(df['slug'])

playerDict = dict(zip(players, slugs))



# Get the Player Data
# You could also make this as a user input

list_of_players = ['lebron james', 'Derik RoSe', 'LEamarkcus Allridge']

data = {}
for player in list_of_players:
    playerMatch = process.extractOne(player, list(playerDict.keys()))[0]
    player_slug = playerDict[playerMatch]

    url = f'https://projects.fivethirtyeight.com/2020-nba-player-projections/{player_slug}.json'
    jsonData = requests.get(url).json()
    
    data[playerMatch] = jsonData
    print(f'Returned: {playerMatch}')

输出：

【讨论】：

Megalina 没有指定使用 Python 请求。 Megalina 希望用 Python Selenium 来完成