CSV to JSON Converter

How this reads your output

The CSV is parsed to RFC 4180, so quoted fields containing commas or newlines survive intact rather than splitting rows. Each header becomes a key, and values that look like numbers or booleans are converted, with two deliberate exceptions: anything with a leading zero and anything longer than fifteen digits stays a string. Those two rules exist because a postcode starting with a zero and a long order number are identifiers rather than quantities.

What the results mean

Values converted from text

How many cells became numbers or booleans. Everything else, including identifiers, stays as strings.

Duplicate column names

A JSON object cannot hold the same key twice, so a duplicate header means the later value silently wins.

Problems in the source

Ragged rows and unterminated quotes in the CSV. These produce misaligned data that looks perfectly valid afterwards.

Common problems and fixes

Leading zeros disappeared in another converter

It converted the value to a number, which is the standard failure for postcodes, product codes and telephone numbers. This converter keeps anything with a leading zero as a string for exactly that reason. If you need everything as strings, that is a safer default for a data pipeline than clever type detection.

Rows are shifted by one column

A quoted field containing the delimiter, split by something that did not respect quoting. The problems count above catches this. Splitting CSV on commas is the most common way to corrupt a file, since a field containing a comma is entirely legal and completely invisible until it breaks something.

Frequently asked questions

More in DevOps

• JSON Formatter / ValidatorFormat and validate JSON — with the error pinned to a line and column.