XML record paths and CSV rows
Automatic selection looks for the first repeated child group. Enter a path such as catalog/item when a document contains more than one repeated section or when you want a specific element. Qualified names can be used in paths. A path with no matching elements produces an error, so an empty export is not mistaken for a successful conversion.
Nested fields, arrays and missing values
An attribute named id becomes @id. An address/city child becomes address.city. Repeated children remain a JSON array inside one CSV cell instead of multiplying rows, which avoids silently creating combinations between unrelated arrays. All records contribute to the header set, and a record without a field receives an empty cell in that column.
CSV delimiters and spreadsheet text
Comma, semicolon, tab and pipe output are available. Fields containing the chosen delimiter, a quote or a line break are quoted, and embedded quotes are doubled. UTF-8 BOM is a download option for spreadsheet import. The textual value 007 remains 007 in the CSV, although a spreadsheet may infer a number when opening it.
Format references
XML 1.0 · CSV fields and quoting
File content is read and converted in the browser. Conversion does not upload the source to a server.