Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

0.18.1

Fixed

Add link to docs in readme

0.18.0

Added

In-depth documentation. Lots of docstring updates!
serdelicacy.SerdeError as part of the public API (mostly for documentation)

0.17.0

Added

Explicit support for enum.Enum and its subtypes. I personally prefer typing.Literal, but enum.Enum might be preferred by others.

0.16.0

Note: breaking changes regarding validation and transformation.

Added

Provide Override type, which provides metadata user-defined overrides for dataclass serde for the namespace-constrained serdelicacy key. Prevents namespace pollution and provides more type safety.

Removed

Keys transform_load, transform_postload, and validate are no longer expected. Instead, use the key serdelicacy and pass it a serdelicacy.Override object.

0.15.0

Added

Support for additional dataclasses.field metadata keys: transform_load, transform_postload, and transform_dump. These enable the user to specify functions to modify the specific value that is deserialized (loaded) and serialized (dumped).

Fixed

Inconsistent documentation the referred to UNDEFINED has been corrected to refer to MISSING.

Removed

Removed automatic conversion for primitive types. This special case created a lot of code complexity and is arguably unnecessary now that we provide support for validations and transformations with dataclasses.

0.14.0

Added

Major validation improvement! Support for dataclass attribute validation using the metadata parameter to dataclasses.field, like dataclasses.field(metadata={"validate": str.istitle}). The function passed as the mapping value should either return True on positive validation / False on non-validation, or return nothing at all and instead rely on the raising of exceptions to indicate whether validation passed for failed.

0.13.0

Added

Support for deserialization of typing.Literal. Lets users to specify an allow-list for string / other input values, removing an entire category of annoying custom validation that users previously needed to write in __post_init__.

0.12.0

Added

New Public API function serdelicacy.get. Similar to dict.get, that returns an item, or its default if the item is MISSING.

Changed

Error messages modified for terseness

0.11.1

Changed

Error message JSON keys are now color encoded to draw visual attention to errors: green/blue for normal, red for the error

0.11.0

Changed

Error messages now in JSON format with better-organized information

Fixed

Resolved bug where str was converted automatically to a List / other sequence. Since str is a special-case primitive, we must handle it as a special case within the _check_sequence method.

0.10.0

Added

is_missing function added to serdelicacy

Changed

UNDEFINED/Undefined changed to MISSING/Missing

0.9.0

Changed

Project name changed from serdataclasses to serdelicacy. This reflects this project's evolution beyond simply dataclasses into NamedTuples, etc. Version 0.9.0 is equivalent to serdataclasses 0.8.0 in every other way.

0.8.0

Major changes in this release!

Added

(potential breaking change) add OptionalProperty, UNDEFINED, and Undefined for missing properties. If your code currently relies on None being generated by typing.Optional on missing properties, read the function signature of load and dump.

Changed

Error messages are generally much better
Better-handle custom exceptions being raised in dataclasses. Useful to elegantly hande validation checks in __post_init__

Fixed

Type improvements for NoResult

0.7.0

Added

Support for automatic conversion of primitive types (str, int, float, bool). Stops the conversion when hitting all non-Option Union types.

0.6.1

Fixed

object types were incorrectly specified, they have been changed to Any.
Update pylint and mypy for local debugging purposes

0.6.0

Added

Support default arguments for NamedTuples and dataclasses. Behavior: if default argument exists, if no argument is present in the unstructure Python object, and if there is a default value for the argument, we use the default argument instead of None (which is automatically returned by dict.get).

0.5.1

There are no functional changes here, just some book-keeping CI / local dev-related stuff.

Added

Set up continuous integration using GitHub Actions.

Fixed

Some development dependencies weren't in pyproject.toml.

0.5.0

Added

load now supports tuple types
dump now supports Sequence/Mapping, not just list/dict

0.4.0

Added

Keywords and classifiers to pyproject.toml.

Changed

Lists and Dicts in deserialize.load have been generalized into the generic protocols typing.Sequence and typing.Mapping.
typing.get_origin and typing.get_args are now call called only once in the Deserialize constructor.

0.3.0

Added

Support for typing.TypedDict
typesafe_constructor argument to deserialize.load to optionally ensure type safety at the top-level. It defaults to True.

0.2.0

Added

This CHANGELOG.md
deserialize.load, deserialize.DeserializeError, and serialize.dump are imported at the top-level.
Support for NamedTuple and Dict
Add py.typed to package (PEP 561)
typedefs.NoResult and typedefs.Possible. This is a bespoke typing.Optional implementation because None is a value we actually care about.

Changed

deserialize.deserialize changed to deserialize.load
serialize.serialize changed to serialize.dump
The deserialize._deserialize function is replaced by deserialize.Deserialize.
Removed "vaporware" designation: this is now a real project.

Fixed

mypy, pylint, black, toml-sort, and isort all pass.

Removed

Recursive types. I may add them again in the future when mypy supports them, but given the sketchyness of the type system at present, I'm not confident that they'll actually help