R语言爬取Google评论时遇Chrome无法访问错误求助

Hey there, let's tackle that frustrating "Chrome not reachable" error you're hitting while trying to scrape Google reviews with RSelenium. I've dealt with this exact issue a handful of times, so here are the most reliable fixes that should get your script back on track:

Common Causes & Solutions #

• Mismatched Chrome and ChromeDriver Versions
  This is the #1 culprit. RSelenium's rsDriver requires your Chrome browser version to match the ChromeDriver version it uses. First, check your Chrome version (go to Chrome Settings > About Chrome). Then, specify the matching ChromeDriver version when initializing rsDriver:

  # Replace with your Chrome version number
  rmDr <- rsDriver(browser = "chrome", chromever = "114.0.5735.90")
  

  If you're unsure which version to use, you can use wdman::chrome_driver() to automatically fetch the correct driver for your installed Chrome.

• Residual Chrome Processes Blocking the Port
  Sometimes leftover Chrome processes (even hidden ones) hog the port RSelenium tries to use. Kill all Chrome instances first:

  • On Windows: Run taskkill /im chrome.exe /f in Command Prompt (or add system("taskkill /im chrome.exe /f", ignore.stdout = TRUE) to your R script)
  • On Mac/Linux: Run pkill chrome in Terminal
    Then restart your R script.

• Port Conflict
  The default port RSelenium uses might be occupied by another app. Try specifying a custom, unused port:

  rmDr <- rsDriver(browser = "chrome", port = 4567L) # Use any number between 1024-65535
  

• Outdated Packages
  Make sure RSelenium, rvest, and xml2 are all up to date. Reinstall them to be safe:

  devtools::install_github("ropensci/RSelenium")
  install.packages(c("rvest", "xml2"))
  

• Try Headless Mode
  Sometimes graphical Chrome has compatibility issues. Switch to headless mode (no browser window pops up) which is often more stable:

  # Set headless Chrome options
  eCaps <- list(chromeOptions = list(args = c('--headless=new', '--disable-gpu')))
  rmDr <- rsDriver(browser = "chrome", extraCapabilities = eCaps)
  

Test Script Example #

Here's a cleaned-up version of your code incorporating these fixes:

library(RSelenium)
library(rvest)
library(xml2)

# Kill residual Chrome processes (Windows example)
system("taskkill /im chrome.exe /f", ignore.stdout = TRUE)

# Initialize driver with matching Chrome version and custom port
rmDr <- rsDriver(browser = "chrome", chromever = "114.0.5735.90", port = 4567L)
myclient <- rmDr$client

# Navigate to the Google reviews page
myclient$navigate("https://www.google.co.uk/search?q=queen%27s+hospital+romford&oq=queen%27s+hospitql+&aqs=chrome.1.69i57j0l5.5843j0j4&sourceid=chrome&ie=UTF-8#lrd=0x47d8a4ce4aaaba81:0xf1185c71ae14d00,1,,,")

# Add a small delay to let the page load fully (anti-scraping best practice)
Sys.sleep(3)

# Get page source
pagesource <- myclient$getPageSource()[[1]]

# Proceed with your SelectorGadget scraping here...

# Don't forget to clean up!
myclient$close()
rmDr$server$stop()

A quick note: Google has strict anti-scraping measures, so avoid sending too many requests in a short time. Adding delays like Sys.sleep() between actions helps prevent getting blocked.

内容的提问来源于stack exchange，提问作者Varun