How CSV, Excel, and XML Conversion Works

All four of PDFPilot's CSV/Excel/XML tools share one XML structure: <rows><row><ColumnName>value</ColumnName>...</row></rows> — one <row> element per spreadsheet row, and one child element per column, named after that column's header. This is a deliberate, generic, documented shape, not a guess at any specific accounting or ERP system's proprietary schema.

Column headers are sanitized into valid XML element names when converting to XML: spaces become underscores, punctuation is stripped, and a name that would start with a digit gets an underscore prefix. A header that sanitizes to nothing usable falls back to a positional name like field_2.

Converting the other direction (XML to CSV or XML to Excel) expects that exact same structure back. Column order and names are taken from the first <row> element — a later row missing one of those fields produces an empty cell rather than shifting the other columns, and XML that doesn't match this shape produces a clear error rather than a guess.

Excel to XML only converts the first sheet with data in a workbook — a real, disclosed limitation, not a bug. A multi-sheet workbook needs a per-sheet output shape this tool doesn't build, since concatenating rows from sheets with different columns would corrupt the data rather than preserve it.

XML to Excel produces a genuine, valid .xlsx file with one sheet — plain data only, with numeric-looking values stored as real numbers and everything else as text. There's no formatting, formulas, or multiple sheets, since the source XML doesn't carry that information either.

As with every tool on PDFPilot, all four conversions happen entirely in your browser. Your file is never uploaded to a server.