TOML & INI to JSON

The parser follows TOML v1.0.0 for the constructs configuration files use: [tables] and nested tables, [[arrays of tables]], dotted keys (a.b.c = 1), basic strings with escapes and literal strings, both multi-line forms, integers with _ separators and 0x / 0o / 0b prefixes, floats including exponents and inf / nan, booleans, arrays (nested and across lines, with trailing commas) and inline tables. Dates and times are kept as strings because JSON has no date type. A duplicate key or a table defined twice is an error with its line number.

INI has no specification, so the page states its conventions: [section] headers, key = value or key: value, ; and # comments (a comment after a value needs whitespace before it), quoted values that keep their spaces, and sections nesting on dots ([app.db] → app → db). With typing on, values that look like integers, decimals or booleans (true / false, yes / no, on / off) become JSON numbers and booleans; switch it off to keep every value a string. Detection looks for TOML-only syntax first ([[…]], arrays, inline tables, dates) and falls back to INI for colon-separated or ;-commented text.

The TOML rule against defining a table after one of its sub-tables, mixed-type array restrictions of TOML 0.x, and the exact date-time grammar (dates are passed through as text). INI dialects with line continuations or %(interpolation)s are read literally. Nothing leaves the browser; the same four anonymous usage counts as the rest of the site apply.