Check and tidy TOML files like Cargo.toml, pyproject.toml and Hugo configs. Paste or drop TOML to validate it against TOML 1.0 with exact error positions, reformat it consistently, or convert it to JSON — and turn JSON into TOML. Several files can be converted at once.
| TOML | JSON |
|---|---|
| 1979-05-27T07:32:00Z | "1979-05-27T07:32:00Z" |
| 0xff | 255 |
| inf | "inf" |
| [[bin]] | "bin": [{…}] |
Paste it — errors show the line and column.
Not when reformatting or converting; the data is preserved.
As ISO 8601 strings, since JSON has no date type.
TOML has no null value. Remove or replace nulls first.
No.
The parser implements TOML 1.0: - **Strings:** basic, literal and both multi-line forms, including line-ending backslashes and \u escapes. - **Integers:** decimal, hex, octal and binary, with underscores only between digits. - **Floats:** exponents, inf and nan. - **Date-times:** offset date-times, local date-times (with T or a space), local dates and local times, with impossible dates rejected. - **Tables:** inline tables, dotted keys, [tables] and [[arrays of tables]]. It enforces the spec's rules — no duplicate keys, no table defined twice, no extending inline tables or static arrays — and reports errors with a line and column. Formatting re-serializes plain keys first, then tables and arrays of tables, optionally sorted; comments are not preserved. For JSON, dates become strings, out-of-range integers keep every digit, and inf and nan become strings.