Skip to contents

Read legacy profile XML into a data frame

xlm_profiles_to_df.Rd

Parses a legacy XML dump where records are stored under <ROW> nodes. The function reads raw bytes, drops NULs, interprets as Latin-1/Windows-1252, converts to UTF-8, fixes/sets the XML declaration, and extracts each <ROW> as a row. Columns are the union of child element names across rows; missing fields are filled with NA.

Usage

xlm_profiles_to_df(path)

Arguments

path

Character scalar. Path to the XML file.

Value

A base data.frame with one row per <ROW> and character columns for each child element name found in the file.

Details

This helper is designed for messy, non-UTF-8 dumps (e.g., Windows-1252) that may contain NUL bytes. It performs minimal normalization and does not coerce types; parse dates or numerics downstream as needed.

See also

xml2::read_xml(), xml2::xml_find_all()

Examples

# \donttest{
xml_txt <- paste0(
  '<?xml version="1.0"?>',
  '<ROWSET>',
  '  <ROW><ID>1</ID><PROFILE_TYPE>buyer</PROFILE_TYPE></ROW>',
  '  <ROW><ID>2</ID><PROFILE_TYPE>seller</PROFILE_TYPE><CITY>Panamá</CITY></ROW>',
  '</ROWSET>'
)
f <- tempfile(fileext = ".xml"); writeLines(xml_txt, f)
xlm_profiles_to_df(f)
#>   ID PROFILE_TYPE   CITY
#> 1  1        buyer   <NA>
#> 2  2       seller Panamá
unlink(f)
# }