CSV to JSON Converter

Delimiter

The tool attempts to detect common delimiters: comma, semicolon, tab and pipe. Detection remains editable. A file containing many commas inside quoted text can confuse automatic detection, and regional exports often use semicolons.

The preview shows the selected delimiter and the detected number of columns before conversion.

Header Row

When the first row contains field names, each following row can become a JSON object.

id,name,status
1,Ada,active
2,Lin,paused

Suggested output:

[
  {"id": "1", "name": "Ada", "status": "active"},
  {"id": "2", "name": "Lin", "status": "paused"}
]

When no header exists, you can output arrays of arrays, generate keys such as column_1 or enter custom field names.

A CSV field may contain the delimiter, a line break or literal quotes when it is quoted correctly.

name,notes
"Ada","Uses commas, quotes, and
multiple lines"

The parser keeps the entire notes value in one field and decodes escaped quote pairs inside quoted values.

Array of Objects

Use the header row as object keys. This is the default for APIs, application imports and most developer workflows.

Array of Arrays

Keep every row as a positional array. This is useful when the CSV has no header or when duplicate headers should not become object keys.

Wrapped Object

An optional root key can produce:

{
  "records": [
    {"id": "1", "name": "Ada"}
  ]
}

The root key must be entered explicitly. The tool does not invent a semantic collection name.

CSV has no universal type metadata. The safest default keeps all cell values as strings.

• Preserve all values as strings — the default, so identifiers such as 00123 keep their leading zeros.
• Detect numbers and booleans — converts numeric text to JSON numbers and true/false to booleans.
• Empty cell handling — keep an empty string or emit null.

An optional dot-path mode converts headers such as address.city and address.country into nested JSON.

{
  "address": {
    "city": "London",
    "country": "UK"
  }
}

This mode is disabled by default because a literal column name can contain a period. Users control whether dots mean nesting.

The preview displays the detected delimiter, header status, row and column count, malformed-row warnings, duplicate or blank header warnings, a small source-table preview and formatted JSON output.

A name containing a comma creates extra fields

The field must be quoted in the source CSV. The parser respects quoted delimiters and reports malformed quoting.

Rows contain different numbers of columns

The file is irregular or the wrong delimiter was selected. The output shows which rows differ; missing values are padded and extra values are kept by default.

Leading zeros disappear

Disable numeric type detection for identifier columns. A value such as 00123 is often an ID, not the number 123.

The first data row became property names

Turn off “first row is header.” Use array output or provide custom keys.

Empty cells should become null

Select the empty-value policy before conversion. The preview distinguishes "" from null.

For large or complex files, start with a representative sample and review the generated output before relying on it.