抓取基本页面时，R 脚本返回一些错误

Question

这是我的脚本：

library(rvest)
library(dplyr)


link = "http://www.mmadecisions.com/decisions-by-judge/"
page = read_html(link)

name = page %>% html_nodes("#page1 a") %>% html_text()
name_links = page %>% html_nodes("#page1 a") %>%
  html_attr("href") %>% paste("http://www.mmadecisions.com/", ., sep="")

get_decisions = function(name_link) {
  judge_page = read_html(name_link)
  date = judge_page %>% html_nodes(".list:nth-child(1)") %>% html_text()
  event = judge_page %>% html_nodes(".list:nth-child(2) a") %>% html_text()
  fight = judge_page %>% html_nodes(".list~ .list+ .list a") %>% html_text()
  decisions = judge_page %>% html_nodes(".list:nth-child(4)") %>% html_text()
  return(judge_page)
}

decision = sapply(name_links, FUN = get_decisions)

judges = data.frame(name, date, event, fight, decision, stringsAsFactors = FALSE)

我不断遇到的错误如下：

> library(rvest)
> library(dplyr)
> 
> 
> link = "http://www.mmadecisions.com/decisions-by-judge/"
> page = read_html(link)
> 
> name = page %>% html_nodes("#page1 a") %>% html_text()
> name_links = page %>% html_nodes("#page1 a") %>%
+   html_attr("href") %>% paste("http://www.mmadecisions.com/", ., sep="")
> 
> get_decisions = function(name_link) {
+   judge_page = read_html(name_link)
+   date = judge_page %>% html_nodes(".list:nth-child(1)") %>% html_text()
+   event = judge_page %>% html_nodes(".list:nth-child(2) a") %>% html_text()
+   fight = judge_page %>% html_nodes(".list~ .list+ .list a") %>% html_text()
+   decisions = judge_page %>% html_nodes(".list:nth-child(4)") %>% html_text()
+   return(judge_page)
+ }
> 
> decisions = sapply(name_links, FUN =  get_decisions)
Error in open.connection(x, "rb") : HTTP error 400.
Called from: open.connection(x, "rb")
Browse[1]> 
> judges = data.frame(name, date, event, fight, score, stringsAsFactors = FALSE)
Error in data.frame(name, date, event, fight, score, stringsAsFactors = FALSE) : 
  object 'score' not found

我的目标是从父页面导航到多个子页面，抓取四列数据“判断决策”，然后打印成列。我感谢任何人对此提出的任何见解。

Answer 1

不完全确定为什么您有两个结束变量。如果您想要这 4 列的最终 df，则可以从每一页使用 purrr::map_dfr，并确保从函数返回 tibble。对于连接问题，您需要修剪您的空白网址。

library(rvest)
library(dplyr)
library(purrr)

link = "http://www.mmadecisions.com/decisions-by-judge/"
page = read_html(link)

name = page %>% html_nodes("#page1 a") %>% html_text()
name_links = page %>% html_nodes("#page1 a") %>%
  html_attr("href") %>% paste("http://www.mmadecisions.com/", ., sep="") %>% trimws()

get_decisions = function(name_link) {
  judge_page = read_html(name_link)
  tibble(
    date = judge_page %>% html_nodes(".list:nth-child(1)") %>% html_text(),
    event = judge_page %>% html_nodes(".list:nth-child(2) a") %>% html_text(),
    fight = judge_page %>% html_nodes(".list ~ .list + .list a") %>% html_text(),
    decisions = judge_page %>% html_nodes(".list:nth-child(4)") %>% html_text()
  ) -> t
  return(t)
}

df <- map_dfr(name_links, get_decisions)