Repeated record-node detection
Candidate repeating elements are listed so the tool never silently decides the wrong row level.
XML row selection and safety checks before CSV export
XML to CSV Converter with Record-Node and Attribute Preview
Convert XML to CSV locally after confirming the repeating record node. Preserve attributes, namespaces, CDATA, repeated values, and leading-zero IDs with DTD and entity-expansion protection.
✓ Runs locally✓ No account✓ Free first release
Start with the exampleLocal browser processingYour file contents are not uploaded.
10 MB · 50,000 row limitConversion workbench
Inspect first. Export second.
Source structure, quality findings, and every deliberate transformation stay visible before download.
1
Source data
Paste XML or upload an XML file
or drop it here
2
Structure choices
Nothing ambiguous is hidden behind “auto.”
Safe XML boundaryDTD and custom ENTITY declarations are rejected before parsing.
CSV export settings
3
Preview and quality report
Analyze the sample or your own data to continue.
Your source data is still only in this browser.
Analysis creates a table preview, a ranked issue list, and an exact transformation summary.
Private by design
Source bytes remain in this browser tab.
Reviewable
Preview values and risks before export.
Traceable
Download a JSON conversion report.
What the converter checks
Quality findings belong beside the output
A successful parse is not proof that the result kept the source meaning. These checks surface the places where tabular conversion can change or hide data.
Attribute and namespace preservation
Attributes use an @ prefix, namespace prefixes remain visible, and collisions are not hidden.
Repeated child strategy
Repeated child elements remain JSON text by default or can be joined when every value is scalar.
Entity-expansion protection
DOCTYPE and custom entity declarations are blocked; built-in XML entities are decoded safely.
Preservation rules
Know what happens to difficult values
The tool records deliberate changes and keeps ambiguous structures visible rather than silently normalizing them.
Read the full check definitions →Source featureConverter behavior
Attributes
Included as columns such as @code instead of being discarded.
Namespace prefixes
Kept in record and column names to avoid accidental collisions.
CDATA
Read as source text without interpreting markup inside it.
Repeated child elements
Stored as JSON arrays by default or joined only on request.
A short, inspectable workflow
From source to checked CSV
- 1
Add the XML
Paste a document or select a local XML file. DTDs are rejected before parsing.
- 2
Confirm the record node
Select the repeating element that should become CSV rows and review nested fields.
- 3
Export the checked CSV
Choose spreadsheet protection and keep the report beside the downloaded CSV.
A concrete failure mode
Why XML needs a record-node decision
An inventory feed may repeat item elements while each item also repeats tag elements. Treating both arrays as rows would change the meaning. DataFormatKit asks which node represents records and keeps nested repetitions visible.
Inspect the included example