CSV to JSON Converter

The first row is always used as the keys, and every row after it becomes one object in a JSON array. The output is indented with two spaces:

Notice that 36 and 41 came out as numbers, without quotes, while the names and cities stayed text. That guess is made separately for every cell, which matters in the next section.

A cell becomes a JSON number whenever JavaScript can read it as one. Otherwise it stays a string. Booleans and null are never produced. Here is what the converter does with some common values:

Three of these catch people out. Codes with leading zeros, like ZIP codes and phone numbers, lose them. A cell holding only a space becomes 0. And putting a value in quotes in the CSV does not protect it: "42" still becomes 42. If you need those columns as text, fix them in the JSON afterwards, or convert them in your own code instead.

Rows don't have to match the header exactly. Missing cells become empty strings, and extra cells past the last header are dropped without a warning:

The word extra is gone. A few header problems also change the keys:

• A blank header becomes column2, column3 and so on, numbered by position.
• A repeated header keeps only the value from its last column.
• A space after a comma stays in the key, so name, city gives you a key called " city" with a leading space.

Tidy the header row before converting and you avoid all three.

Only commas separate fields. A semicolon-separated export comes out as one long key such as "name;age" holding the whole row. Re-export with commas, or replace the semicolons first if no value contains one.

Windows (CRLF) and Unix (LF) line endings both work, blank lines are skipped, and a line break inside a quoted cell is kept as \n in the string.