JSON to CSV Converter

JSON to CSV Converter turns records into one table, including nested fields.

Use this tool

Output CSV format

Parsed table

Result

The result appears in this panel after you run the tool.

About this tool

Include header row writes the flattened column names, then one row per record. Turn it off to write only the data rows. Nested objects use dotted paths, and arrays use indexes such as tags[1]. A missing key is an empty cell. A value that contains the delimiter is quoted.

CSV to JSON Converter reads a CSV table back into JSON objects.

How to use it

  1. Paste JSON. A list of records, a nested list, a map of records, or one object can become the table.
  2. If several row collections are equally valid, choose Rows from. Then set the output delimiter and header.
  3. Convert to CSV. Copy, download, or continue with that serialized table.

Limitations

Records flattened into cells

A table needs records, not a bare list of scalars

The converter looks for a collection of objects, or a map of objects, and turns that collection into rows. A JSON array of numbers or strings has no records, and the run stops rather than inventing a column per scalar. A single object can become one row. When several collections are equally plausible, you choose which path supplies the rows. Nested lists inside the chosen records are not written as extra files. They are folded into the row.

Folding is the structural mismatch. An object under a field becomes dotted column names, so a role object with a title and a rank becomes columns for those paths rather than a cell that still contains an object. A list of scalars becomes numbered columns starting at one, so two tags occupy a first and a second column instead of one cell holding both. Null and missing fields become empty cells. Numbers and booleans become text in those cells. The grid cannot remember that a cell used to be null rather than an empty string, or that a number used to be a number.

The destination dialect decides the file, not the JSON

You choose the output separator, quote, escape, and whether a header row of those column paths is written. Turning the header off drops the names from the file. The cells remain, and a later reader will not know that the third column was a dotted path unless you kept the header. A separator of spaces is written as a comma, because the writer has no multi-character space delimiter. Quotes are added when a cell contains the separator, a quote, or a break.

The on-screen grid can be shorter than the result

The preview draws a limited number of rows and columns and says so when it clips. The text you copy is the full grid. Empty input and a shape with no records fail. Use this page to project records onto a rectangle. Do not use it when the nesting itself is the data you must keep. YAML and XML converters that target CSV share the same flattener, with their own parsers in front.

Privacy

Your input is processed in your browser and is not uploaded to UtilChain. File names, text, and results are not sent to analytics.