skip error Error in open.connection(x, "rb") : HTTP error 404

Question

Hello, Im new to this fascinating world of r, I have not been able to skip the urls that do not exist, how can I handle it? and don't mark as and error, thanks for your help.

---

title: "error" author: "FJSG" date: "27/6/2020" output: html_document

knitr::opts_chunk$set(echo = TRUE)

library(xml2)
library(rvest)
library(tidyverse)
library(lubridate)

zora_core <- read_html("https://zora.medium.com/the-zora-music-canon-5a29296c6112")

Los_100 <- data.frame(album      = html_nodes(zora_core, "h1:not(#96c9)") %>% 
                                     html_text() %>% 
                                     str_trim(side = "both"),
                      interprete = html_nodes(zora_core, "strong em , p#73e0 strong") %>% 
                                     html_text() %>% 
                                     str_remove_all("^by") %>%
                                     str_extract("[a-zA-Z].+(?=[(])") %>% str_trim(side = "both"),
                      año        = html_nodes(zora_core, "strong em , p#73e0 strong") %>% 
                                     html_text %>% 
                                     str_extract("([[:digit:]]){4}"),
                      liga       = paste0("https://en.wikipedia.org/wiki/",html_nodes(zora_core,                                       "strong em , p#73e0 strong") %>% 
                                     html_text() %>%
                                     str_remove_all("^by") %>%
                                     str_extract("[a-zA-Z].+(?=[(])") %>% str_trim(side = "both") %>% str_replace_all(" ","_")))

carga <- function(url){
  
         perfil_raw <- read_html(url)
         data.frame(interprete = html_node(perfil_raw, "h1#firstHeading") %>% 
                                 html_text() %>% str_trim(side = "both"))
         
}

lista <- Los_100$liga[1:16] # THE url for the position 16 don´t exist how to avoid that

datos_personales <- map_df(lista,carga)

Answer 1

It's useful to learn about error-handling in R, but when working with http requests it becomes essential.

In your case, it is best to wrap carga in a tryCatch. This runs an expression that you pass as the first argument and if an error is thrown, it is caught and passed to the second argument of tryCatch, which is a function.

If an error is thrown we need to return a data frame with a single column called interprete so that map_df can bind it together with the other results:

carga_catch <- function(x)
{
  tryCatch(return(carga(x)),
           error = function(e) return(data.frame(interprete = "**inexistente**")))
}

map_df(lista, carga_catch)
#>               interprete
#> 1        Ella Fitzgerald
#> 2          Sarah Vaughan
#> 3         Billie Holiday
#> 4  Sister Rosetta Tharpe
#> 5             Lena Horne
#> 6        Mahalia Jackson
#> 7          Abbey Lincoln
#> 8             Etta James
#> 9         Leontyne Price
#> 10       Marian Anderson
#> 11      Dinah Washington
#> 12                Odetta
#> 13        Dionne Warwick
#> 14          The Supremes
#> 15           Nina Simone
#> 16       **inexistente**

Apart from error handling, I think your code is very good for someone just beginning in R. It achieves a lot in a few lines of code and is perfectly readable. Good work!