Save `xml_document/xml_node` object in a tibble for mutating-CodePudding

I would like to save a html page to a tibble so that I could later use mutate on the page content

I thought about reading the html direct to a tibble:

library(tidyverse)
library(rvest)

#does not work
tibble(html=read_html("https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=040445"))
#> Error: All columns in a tibble must be vectors.
#> x Column `html` is a `xml_document/xml_node` object.

Reading as a list works:

#works
works <- tibble(html=list(read_html("https://www.accessdata.fda.gov/scripts/cder/daf/index.cfm?event=overview.process&ApplNo=040445")))
works
#> # A tibble: 1 x 1
#>   html      
#>   <list>    
#> 1 <xml_dcmn>

However, I cannot use mutate then:

# does not work
works %>% 
  mutate(table=html_nodes(unlist(page),"#exampleApplSuppl"))
#> Error: Problem with `mutate()` column `table`.
#> i `table = html_nodes(unlist(page), "#examleApplSuppl")`.

^{Created on 2021-11-02 by the reprex package (v2.0.1)}

CodePudding user response：

As the 'html' column is list, loop over the list and return the output in a list

library(purrr)
library(dplyr)
works %>% 
   mutate(table = map(html, ~ html_nodes(.x, "#examleApplSuppl")))

-output

# A tibble: 1 × 2
  html       table     
  <list>     <list>    
1 <xml_dcmn> <xml_ndst>