changelog.rst

Changelog

v2.11

• fixed a bug with serializing large integers
• invalid UTF-8 field name will now display name of the field as repr
• major bugfix: UTF-8 characters that took more than 1 bytes were serialized wrong

v2.10

• fixed a bug in which if :meth:`minijson.MiniJSONEncoder.default` returned a bytes object or a boolean, it would be classified as an invalid value.

v2.9

• minor refactor: code deduplication
• fixed some bugs with unserializing ints on other platforms than x86_64

v2.8

• serializing a dict without use_strict_order won't construct a list of it's items,

  allowing you to serialize large dictionaries

v2.7

• added option to sort dictionary keys before serialization

v2.6

• added option to serialize binary data

v2.5

• added :class:`minijson.MiniJSONEncoder`

v2.4

• added argument default
• fixing issue with serializing classes that subclass dict, list and tuple

v2.3

• :func:`minijson.loads` will now take any object that can provide it's __bytes__

v2.2

• added support for PyPy and Python 3.5

v2.1

• proofed against loading empty strings
• Python 3.6 is supported
• minor speed improvements

v2.0

• fixed a bug with serializing uint32a
• added support for arbitrarily large integers
• major refactor
• backwards compatible 100%

v1.5

• fixed a bug with wrong type of dict and string was chosen

  for a dict which contains exactly 65535 keys. Since this is rare in production, it can wait. MiniJSON is still generated correctly.

• fixed a bug with dumping strings longer than 255 characters

  would not return a length

• fixed a bug with unserializing some strings

v1.4

• more compact representation for not-all-keys-are-strings object

v1.3

• object keys don't have to be strings anymore

v1.2

• removed the limit for string length and list and object size

v1.1

• fixed to work under older Pythons (got rid of the f-strings)
• fixed docstrings to signal that some functions raise exceptions
• fixed a bug with encoding long lists

v1.0

• first release