XML to CSV Converter

XML to CSV conversion cannot be reliable until the tool knows what one record is. The record selector is therefore a primary control, not an advanced afterthought.

Consider this XML:

<orders>
  <order id="1001">
    <customer>Ada</customer>
    <total>25.00</total>
  </order>
  <order id="1002">
    <customer>Lin</customer>
    <total>40.00</total>
  </order>
</orders>

The repeated path /orders/order becomes two CSV rows.

• scan for repeated sibling elements
• suggest likely record paths with their counts
• let you choose a different node
• preview the candidate records before conversion

Nested elements can become path-based columns.

<customer>
  <name>Ada</name>
  <address>
    <city>London</city>
  </address>
</customer>

Suggested columns:

customer.name
customer.address.city

The options panel includes flatten nested elements, the path separator, the maximum depth, the behavior after the maximum depth and how to preserve a nested fragment.

Attributes often contain IDs, types or metadata that should become columns.

<order id="1001" status="paid">

Recommended columns:

@id
@status

• include all attributes
• exclude attributes
• choose the attribute prefix
• keep or strip namespace prefixes

A record may contain a repeated child array:

<order>
  <tag>priority</tag>
  <tag>international</tag>
</order>

• Join values with a selected separator.
• Keep the XML fragment in one quoted cell.

XML namespace prefixes can distinguish elements that otherwise have the same local name.

By default, prefixes are stripped from column names. If two different namespaced elements produce the same local name, a collision warning is shown and prefixes are preserved for the affected columns.

API capabilities determine input and table limits. Previews may be truncated; download the full output artifact.

After XML records are selected and flattened, ordinary CSV output controls apply: delimiter, include header, minimal or all-field quoting, line ending, empty and missing value behavior and UTF-8 output.

Values containing delimiters, quotes or line breaks are escaped correctly.

Before export, the preview shows the selected record path, the number of records found, the generated columns, warnings for fields missing from some records, repeated-child warnings, namespace collisions and a table preview.

The converter created only one row

The root element was selected instead of the repeated record node. Choose the child path that appears once for each record.

Attributes are missing

Enable attribute columns. The preview shows the chosen prefix and any name collisions.

Nested values disappeared

Increase the flatten depth or preserve the nested fragment. The tool warns when data exists beyond the configured depth.

Several child values appear in one cell

The record contains a repeated child list. Choose join or preserve behavior.

Column names are duplicated

Namespace prefixes or attribute markers may have been removed. Use a naming strategy that keeps each source path unique.

For large or complex files, start with a representative sample and review the generated output before relying on it.