CODEWITHSUNDEEP
rxml
user4394417
user4394417

Reputation:

R - extract part of xml with condition

I have this dataframe. What I'm looking for is to extract id value od provider if name is goldman. Please notice that in some rows is no goldnam provider so result should be NA.

df <-
  data.frame(
    id = c(1, 2, 3),
    xml = c(
      as.character(
        "<?xml version=\"1.0\" encoding=\"UTF-8\"?>
          <response>
            <provider name=\"bank_of_ammerica\">
                <success>true</success>
                <id>12</id>
            </provider>
            <provider name=\"goldman\">
                <success>true</success>
                <id>13</id>
            </provider>
          </response>",
        
        "<?xml version=\"1.0\" encoding=\"UTF-8\"?>
          <response>
            <provider name=\"bank_of_ammerica\">
                <success>true</success>
                <id>12</id>
            </provider>
            <provider name=\"goldman\">
                <success>true</success>
                <id>16</id>
            </provider>
          </response>",
        
        "<?xml version=\"1.0\" encoding=\"UTF-8\"?>
          <response>
            <provider name=\"bank_of_ammerica\">
                <success>true</success>
                <id>12</id>
            </provider>
          </response>"
      )
    )
  )

So result should be

result <-
  data.frame(
    id = c(1:3),
    id_val = c(13, 16, NA_integer_)
  )

Upvotes: 0

Views: 55

Answers (1)

Onyambu
Onyambu

Reputation: 79208

aha. Remove the command as.character. it messes with everything.

Anyway. You could do:

library(tidyverse)
library(rvest)
library(xml2)
df%>%
  mutate(id_Val = map_chr(xml,~as_xml_document(.x)%>%
               html_node("provider[name=goldman] id")%>%
               html_text()))%>%
  select(-xml)

  id id_Val
1  1     13
2  2     16
3  3   <NA>

Upvotes: 1

Related Questions