Merge pull request #41 from PainterQubits/#40-make-kw_only-true-by-de…

…fault

#40 make kw only true by default

CHANGELOG.md

@@ -11,12 +11,12 @@ project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
 - Ability to specify start and end indices in `ParamDB.commit_history()`
 - Support for scalar [`astropy.units.Quantity`] objects
+- Parameter dataclass bases (`Param` and `Struct`) automatically convert subclasses into
+  dataclasses `kw_only` as True by default
 
 ### Changed
 
 - `ParamDict` can be initialized from keyword arguments in addition to dictionaries
-- Parameter dataclass bases (`Param` and `Struct`) automatically convert subclasses into
-  dataclasses, so the `@dataclass` decorator should no longer be used
 
 ## [0.2.0] (Mar 8 2023)
 

README.md

@@ -35,6 +35,8 @@ poetry source add --secondary paramdb https://painterqubits.github.io/paramdb/re
 Then the package can be installed like any other (e.g. `poetry add paramdb`).
 
 [poetry]: https://python-poetry.org
+[astropy]: https://docs.astropy.org/en/stable/index.html
+[`astropy.units.quantity`]: https://docs.astropy.org/en/stable/api/astropy.units.Quantity.html#astropy.units.Quantity
 
 <!-- end installation -->
 
@@ -50,8 +52,6 @@ ParamDB has two main components:
 
 See the [api reference] for more information.
 
-[astropy]: https://docs.astropy.org/en/stable/install.html
-[`astropy.units.quantity`]: https://docs.astropy.org/en/stable/api/astropy.units.Quantity.html#astropy.units.Quantity
 [**parameter data**]: https://painterqubits.github.io/paramdb/parameter-data.html
 [**database**]: https://painterqubits.github.io/paramdb/database.html
 [api reference]: https://painterqubits.github.io/paramdb/api-reference

docs/parameter-data.md

@@ -53,15 +53,18 @@ param.value
 
 The dataclass aspects of the subclass can be customized by passing keyword arguments when
 defining the custom class (the same arguments that would be passed to the [`@dataclass`]
-decorator), and by using the dataclass [`field`] function. For example,
+decorator), and by using the dataclass [`field`] function. The class arguments have the
+same default values as in [`@dataclass`], except `kw_only` is True by default for
+ParamDB dataclasses to facilitate dataclass inheritance with default values. An example of
+dataclass customization is shown below.
 
 ```{jupyter-execute}
 from dataclasses import field
 
-class CustomizedDataclassParam(Param, kw_only=True, repr=False):
+class CustomizedDataclassParam(Param, kw_only=False, repr=False):
     values: list[int] = field(default_factory=list)
 
-customized_dataclass_param = CustomizedDataclassParam()
+customized_dataclass_param = CustomizedDataclassParam([1, 2, 3])
 customized_dataclass_param.values
 ```
 
@@ -97,6 +100,12 @@ param_custom_init = ParamCustomInit()
 ```
 ````
 
+```{tip}
+Since the base class of all parameter classes, {py:class}`ParamData`, is an abstract class
+that inherits from [`abc.ABC`], you can use abstract decorators in parameter and structure
+classes without inheriting from [`abc.ABC`] again.
+```
+
 Parameters track when any of their properties was last updated in the read-only
 {py:attr}`~Param.last_updated` property. For example:
 
@@ -170,7 +179,7 @@ example,
 ```{jupyter-execute}
 from paramdb import ParamList
 
-param_list = ParamList([CustomParam(1), CustomParam(2), CustomParam(3)])
+param_list = ParamList([CustomParam(value=1), CustomParam(value=2), CustomParam(value=3)])
 param_list[1].parent == param_list
 ```
 
@@ -233,4 +242,5 @@ This does nothing to the functionality, but static type checkers will now know t
 [`__init__`]: https://docs.python.org/3/reference/datamodel.html#object.__init__
 [`@property`]: https://docs.python.org/3/library/functions.html#property
 [`__post_init__`]: https://docs.python.org/3/library/dataclasses.html#post-init-processing
+[`abc.abc`]: https://docs.python.org/3/library/abc.html#abc.ABC
 [`collections.abc`]: https://docs.python.org/3/library/collections.abc.html

paramdb/_param_data/_dataclasses.py

@@ -9,14 +9,14 @@
 from paramdb._param_data._param_data import ParamData
 
 
-@dataclass_transform()
+@dataclass_transform(kw_only_default=True)
 class _ParamDataclass(ParamData):
     """Base class for parameter dataclasses."""
 
-    def __init_subclass__(cls, /, **kwargs: Any) -> None:
+    def __init_subclass__(cls, /, kw_only: bool = True, **kwargs: Any) -> None:
         # Convert subclasses into dataclasses
         super().__init_subclass__()
-        dataclass(cls, **kwargs)
+        dataclass(kw_only=kw_only, **kwargs)(cls)
 
     def __getitem__(self, name: str) -> Any:
         # Enable getting attributes via indexing