txt

Plain text

The txt codec is for decoding/encoding from/to plain text. It's main purpose is to provide a decoder for the text/plain media type (in particular for decoding outputs of Jupyter Notebook cells which have this media type). It may also be useful for decoding user input in plain text form fields.

Decoding

The txt codec only decodes content to primitive node types i.e. null, boolean, number and string. It makes no attempt to decode more complex structure or semantics of the content.

The string "null" is decoded to null, the string "true" and "false" are decoded to boolean values, and strings with only digits and an optional decimal place are decoded to a number.

Everything else is decoded to a string. For example, this plain text content (with some Unicode characters),

Yay, plain text! 👏

is decoded into a string, encoded in JSON5 as,

'Yay, plain text! 👏'

Encoding

All Node types are encoded as plain text. For non-string primitive nodes, e.g. boolean, number, the plain text representation comes from the toString() method.

No attempt is made to perform lossless encoding of more complex nodes. For Array nodes, items are encoded and joined by a single space. For Object nodes, if there is a content node then that is encoded only, otherwise values are encoded as space separated strings. For example, the following Article node, encoded as JSON5,

{
  type: 'Article',
  title: 'My simple article',
  authors: [
    {
      type: 'Person',
      givenNames: ['Peter'],
      familyNames: ['Pan']
    }
  ],
  content: [
    {
      type: 'Paragraph',
      content: ['My first paragraph.']
    }
  ]
}

is encoded in txt as,

My first paragraph.

Invertability

The txt codec is only invertible for primitive node types. It is lossy for array, object and other more complex node types because it intentionally makes no attempt to encode these as plain text. Use one of the data transfer formats e.g. json or yaml for lossless text-based encoding.