r - 将值迭代地附加到循环创建的数据帧中的所有记录

Question

试图从basketball-reference.com(正在运行)中抓取球员职业生涯中的个人比赛统计数据，但我想将球员姓名添加到与个人比赛结果相对应的结果df中。例如，第一个循环只会为刮取生成的 86 行重复“Kareem Abdul-Jabbar”86 次。我正在尝试使用 cbind 填充方法将下一个循环添加到名为“Player_Name”的现有列中，但 cbind 改为为每个循环创建一个新列。任何关于如何将球员姓名放入单列的建议将不胜感激。

library(rvest)
library(dplyr)

# Create df of players to be scraped
#########################################################################
players = data.frame(player_name = c(rep("Kareem Abdul-Jabbar",each=20),
                                rep("Karl Malone",each=19)),
                     player_id = c(rep("abdulka01",each=20),
                                rep("malonka01",each=19)),
                     initial = c(rep("a",each=20),
                                 rep("m",each=19)),
                     year = c(seq(1970,1989,by=1),
                              seq(1986,2004,by=1)))

# Scrape data and stack in a df
#########################################################################
output <- data_frame()
for (i in 1:2){
  
  url <- paste0("https://www.basketball-reference.com/players/",
                players[i,3],"/",players[i,2],"/gamelog/",players[i,4])
  
  webpage <- read_html(url)
  
  temp <- webpage %>%
    html_nodes("#pgl_basic") %>%
    html_table()
  
  player_name=players[i,1]
  
  output <- cbind(bind_rows(output, temp),player_name)
}

Answer 1

您可以创建 URL 来抓取和使用 map_df将它们组合成一个数据帧。

library(rvest)
library(tidyverse)

urls <- sprintf("https://www.basketball-reference.com/players/%s/%s/gamelog/%s", 
        players$initial, players$player_id, players$year)

result <- map_df(urls, ~.x %>% 
                  read_html() %>%
                  html_nodes("#pgl_basic") %>%
                  html_table(), .id = 'playername') %>% 
  mutate(playername = players$player_name[as.numeric(playername)])