CSV to JSON Converter Guide
Reference for converting CSV rows into JSON arrays, including supported input, output shape, browser-side processing behavior, limits, and common errors.
CSV dialect and output contract
CSV to JSON Converter converts pasted delimited text. Choose the delimiter and whether the first row is a header. With headers, each data row becomes a JSON object. Without headers, each row becomes an array. Cell values remain strings: 001, false and an empty cell do not automatically become a number, Boolean or null.
Common errors: quotes, line breaks and row length
A quoted cell can contain the delimiter or a line break. Two quotes inside a quoted cell represent one literal quote. For example, "Growth, Ops" is one cell, and "say ""hello""" becomes say "hello".
An unmatched quote, unexpected text after a closing quote, or inconsistent column count prevents export. The error includes position or row information where available. Do not add or delete delimiters blindly: inspect quoted cells first. A UTF-8 BOM at the start is removed with a warning. Replacement characters indicate that encoding may already have been damaged before pasting.
Header naming and nested keys
Blank headers receive names such as column_2. Duplicate headers receive a unique numbered suffix and a warning. Existing names are reserved: a,a,a_2 becomes a,a_3,a_2, so no value is silently overwritten. Compare the final keys with the receiving schema instead of relying on suffixes as permanent business field names.
Nested-key mode interprets dots as object paths. user.name can become a nested user object. Overlapping scalar/object paths such as user,user.name are rejected because the same value cannot simultaneously be text and an object. Reserved prototype-related path segments are rejected in nested mode. Keep nested mode off when dots are literal column-name characters.
Review before sharing
Use the CSV fixture case to reproduce a quoted comma, a quoted newline, leading-zero IDs and a duplicate-header collision. The output is not anonymized and has not been schema-validated. Inspect string types with JSON Formatter and compare revisions with Diff Checker.
The converter processes pasted text locally. Large files still consume browser memory; use a reproducible import script for datasets too large to inspect comfortably. See Data handling for site-level processing and analytics boundaries.