XML to CSV: flattening nested data into rows and columns

XML documents are trees; CSV is a flat grid of rows and columns. Converting from one to the other means answering two questions a spreadsheet doesn't have to: which repeated element becomes a row, and how does a nested structure inside that element become a column name?

Finding "the rows"

The first step is identifying which element in the document repeats — that's what becomes one row per occurrence. A document like a product catalog, an order export, or an RSS feed usually has one obvious repeating element (<product>, <order>, <item>); a converter finds the first element that appears more than once at the same level and treats each occurrence as a row.

Flattening nested paths into column names

Once the row element is chosen, anything nested inside it needs a single flat column name. The standard approach is dot notation: <address><city>Berlin</city></address> nested inside a row element becomes a column literally named address.city. Repeated child elements inside a row get an index: a second <tag> element becomes tags[1] alongside tags[0] for the first. This is mechanical and lossless, but it does mean deeply nested XML produces wide, awkward-looking CSV — that's an inherent tradeoff of flattening a tree into a grid, not a bug in any particular converter.

Ragged data is normal, not an error

Not every row will have every possible column — an optional <discount> element that only appears on some products, for example. A correct converter takes the union of every column name that appears across all rows and leaves cells blank where a given row's XML didn't include that element, rather than erroring out on the first row that's missing a field later rows have.

Our XML to CSV converter auto-detects the repeating element and flattens nested paths using dot notation, entirely in your browser — worth a quick look at the column headers on a sample before building a pipeline around the exact shape it produces for your specific document.