XML

library(xml2)
library(purrr)
library(dplyr)

## 
##  次のパッケージを付け加えます: 'dplyr'

##  以下のオブジェクトは 'package:stats' からマスクされています:
## 
##     filter, lag

##  以下のオブジェクトは 'package:base' からマスクされています:
## 
##     intersect, setdiff, setequal, union

# Modified XML data
xml_str <- "
<root>
  <person>
    <name>John</name>
    <age>30</age>
    <address_SET>
      <address>
        <city>New York</city>
        <state>New York</state>
      </address>
    </address_SET>
  </person>
  <person>
    <name>Alice</name>
    <age>253</age>

    <address_SET>
      <address>
        <city>Los Angeles</city>
        <state>California</state>
      </address>
      <address>
        <city>San Francisco</city>
        <state>California</state>
      </address>
      <address>
        <city>Osaka</city>
        <state>Osaka</state>
      </address>
    </address_SET>
  </person>
  <person>
    <name>Bob</name>
    <age>28</age>
    <address_SET>
      <address>
        <city>Tokyo</city>
        <state>Tokyo</state>
      </address>
      <address>
        <city>CIIII</city>
        <state>Saitama</state>
      </address>
      <address>
        <city>Chiba</city>

      </address>
      <address>
        <city>Ibaraki</city>
        <state>Mito</state>
        <state2>Mito</state2>

      </address>
    </address_SET>
  </person>
</root>"

# Parse the modified XML data
xml_data <- read_xml(xml_str)

xml_file <- xml_str
root_element <- "//person"
key_element <- ".//name"
child_element <- ".//address_SET/address"
required_elements <- c("city", "state","state2")

process_xml_data_4 <- function(xml_file, root_element, key_element, child_element, required_elements, search_method = "auto") {
  # Parse the XML data
  xml_data <- read_xml(xml_file)
  
  target_data <- xml_find_all(xml_data, root_element) %>%
    map(~ {
      key_data <- xml_text(xml_find_first(.x, key_element))
      child_data_set <- xml_find_all(.x, child_element)
      child_data <- map(child_data_set, ~ {
        element_data <- map(required_elements, function(y) xml_text(xml_find_first(.x, paste0(".//", y))))
        as_tibble(setNames(element_data, required_elements))
      })
      
      tibble(
        key_data = key_data,
        child_data = child_data
      )
    })
  
  target_data
  result_df <- bind_rows(target_data)
  result_df %>% tidyr::unnest(child_data)
}
result_data <- process_xml_data_4(xml_file, root_element, key_element, child_element, required_elements)
print(result_data)

## # A tibble: 8 × 4
##   key_data city          state      state2
##   <chr>    <chr>         <chr>      <chr> 
## 1 John     New York      New York   <NA>  
## 2 Alice    Los Angeles   California <NA>  
## 3 Alice    San Francisco California <NA>  
## 4 Alice    Osaka         Osaka      <NA>  
## 5 Bob      Tokyo         Tokyo      <NA>  
## 6 Bob      CIIII         Saitama    <NA>  
## 7 Bob      Chiba         <NA>       <NA>  
## 8 Bob      Ibaraki       Mito       Mito