使用现有数据框循环读取 API 响应

Question

我有一个数据框：

df

                        NAME      ARTISTNAME                    COL3
1 Everything_Now (continued)     Arcade Fire Everything_Now%20(continued)%20Arcade%20Fire
2             Everything Now     Arcade Fire             Everything%20Now%20Arcade%20Fire
3              Signs of Life     Arcade Fire            Signs%20of%20Life%20Arcade%20Fire
4           Creature Comfort     Arcade Fire           Creature%20Comfort%20Arcade%20Fire
5                  Peter Pan     Arcade Fire                  Peter%20Pan%20Arcade%20Fire
6                  Chemistry     Arcade Fire                    Chemistry%20Arcade%20Fire

我的目标是用 Genius Lyric 的 API 循环这个，以获取 COL3 中每个值的歌词 url。

如果我不循环播放，只为每首歌曲单独播放，那么我的输出将如下所示：

genius_url <- "https://api.genius.com/search?q=Everything_Now%20(continued)%20Arcade%20Fire"
getgeniuslyrics <- GET(genius_url, add_headers(Authorization = HeaderValue))
geniuslyrics <- jsonlite::fromJSON(toJSON(content(getgeniuslyrics)))
answer <- data.frame(geniuslyrics$response$hits$result$url[1])
answer

  X.https...genius.com.Arcade.fire.everything.now.continued.lyrics.
1    https://genius.com/Arcade-fire-everything-now-continued-lyrics 

str(answer)

'data.frame':   1 obs. of  1 variable:
$ X.https...genius.com.Arcade.fire.everything.now.continued.lyrics.: Factor w/ 1 level "https://genius.com/Arcade-fire-everything-now-continued-lyrics": 1

到目前为止，这是我对 for 循环的尝试，但出现错误：

for(i in 1:length(df[,3])) {
  genius_url <- paste("https://api.genius.com/search?q=", 
                      df3[i,3], 
                      sep="")
  getgeniuslyrics <- GET(genius_url, add_headers(Authorization = HeaderValue))
  geniuslyrics <- jsonlite::fromJSON(toJSON(content(getgeniuslyrics)))
  answer <- data.frame(geniuslyrics$response$hits$result$url[1])
  df[i,4] <- answer[1,]
}

我得到的错误信息是：

Error in x[...] <- m : replacement has length zero
In addition: There were 26 warnings (use warnings() to see them)

希望这是有道理的。任何帮助都会很棒，谢谢。

Answer 1

您的数据框是否已经包含第三列，或者您要从第 1 列和第 2 列创建它？我假设您必须根据第一列和第二列创建第三列。

尝试以类似格式的函数重写一次试验：

 funfun <- function(...){
  x=unlist(list(...))
  A=paste(unlist(lapply(x,strsplit," ")),collapse = "%20")
  genius_url=paste0("https://api.genius.com/search?q=",A)
  getgeniuslyrics <- GET(genius_url, add_headers(Authorization = HeaderValue))
  geniuslyrics <- jsonlite::fromJSON(toJSON(content(getgeniuslyrics)))
  answer <- data.frame(geniuslyrics$response$hits$result$url[1])
  answer
  }

nom 也许你可以从这里循环或使用应用函数：

  apply(df[,1:2],1,funfun)

如果你有第三列，那么你的生活会更轻松：

  funfun_1 <- function(x){
    genius_url=paste0("https://api.genius.com/search?q=",x)
    getgeniuslyrics <- GET(genius_url, add_headers(Authorization = HeaderValue))
    geniuslyrics <- jsonlite::fromJSON(toJSON(content(getgeniuslyrics)))
    answer <- data.frame(geniuslyrics$response$hits$result$url[1])
    answer
  }
sapply(df[,3],funfun_1)