| Commit message (Collapse) | Author | Age | Files | Lines |
| | |
|
| | |
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In this very large commit we tackle a few aspects of Pint that
makes it difficult to do static typing.
1. Dynamic classes became static: Quantity and Unit are now
(for the most part) static classes with a static inheritance.
This allows mypy/pylance and other type checker to properly
inspect them.
2. Added types through out all the code. (WIP)
3. Refactor minor parts of the code to make it more typing
homogeneous. Catch a few potential bugs in the way.
4. Add several TODOs that need to be addressed in 0.23
5. Moved some group and system and context code out of the PlainRegistry
6. Moved certain specialized methods out of the PlainRegistry.
|
| | |
|
| | |
|
| |
|
|
|
|
|
|
|
|
|
| |
While there is still a lot of work to do (mainly in Registry, Quantity, Unit),
this large PR makes several changes all around the code. There has not been
any intended functional change, but certain typing improvements required
code minor code refactoring to streamline input and output types of functions.
An important experimental idea is the PintScalar and PintArray protocols,
and Magnitude type. This is to overcome the lack of a proper numerical
hierarchy in Python.
|
| |
|
|
| |
static to ease with typing
|
| | |
|
| | |
|
| | |
|
| | |
|
| | |
|
| |\ |
|
| | | |
|
| | |\ |
|
| | | |\ |
|
| | | | |
| | | |
| | | |
| | | | |
before and after brackets, e.g. 2 (3 + 4) -> 2 * (3 + 4)
|
| | | | |
| | | |
| | | |
| | | | |
- add tests for pre-processed tree evaluation
|
| | | |\ \
| | | |/ |
|
| | | | |
| | | |
| | | |
| | | |
| | | |
| | | |
| | | | |
Fixes #1621
I've also modified conftest, to facilitate testing with customized unit registries.
This is to be discussed, obviously.
|
| | | |/
| |/|
| | |
| | |
| | |
| | |
| | |
| | | |
Currently, as tokenize.tokenizer does not parse ° as a NAME,
it is translated to " degree". However, the preceding space
means that units with a ° in the middle, for example, Δ°C,
cannot be parsed. Removing the preceding space fixes this,
and does not appear to break any tests.
|
| | | | |
|
| |/ / |
|
| |/ |
|
| |\
| |
| | |
Final step to split the registry from the parser
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Overview:
- All the code in facets is now independent of the definition textual format.
In particular, defintions such as UnitDefinition, ContextDefinition an so on
cannot be built directly from a string.
(some functions are kept only temporarily to simplify but transition)
Building Definition objects from string requires a parser that emits them.
- The standart pint format is implemented in delegates/txt_parser
using flexparser. Briefly each single line statement is mapped to
a ParsedStatement class and each larger construct to a Block class.
- The registry then has an adder function that takes a definition an
incorporate it into the registry.
A few nice features of this approach:
1. The Definition objects are standalone public objects,
you can now build them in a programatic way and incorporate
them to the registry using the define function that will
dispatch to the correct adder:
>>> new_unit = UnitDefintion( ....)
>>> ureg.define(new_unit) # might be called add in the future
No more being forced to use string definitions
(but you can still use them if you want)
2. Composition over inheritance. The Registry does not know how to
parse a definition, but it delegates this to another class which
can be changed. This makes it very easy to write another parser
(faster, simpler) o try out a completely different file format.
3. Error messages can be more meaningful.
Backwards incompatible changes
- is_base parameter Definitions is not needed any more. It is
now computed automatically leading to a leaner experience and
also avoiding incompatible states
- alias for dimensionality has been removed (for now at least)
The only one defined was speed as an alias of velocity.
- (Context|Group|System).from_lines and Definition.from string
have been rewritten in terms of the new parser.
But will be likely removed in the future
- Changing non_int_type is not possible after registry has been
created
- load_definition raises FileNotFoundError instead of a generic exception
if the file was not found
- the string representation of several definitions is now
not so user friendly terms of the new parser.
But will be likely removed in the future
- Changing non_int_type is not possible after registry has been
created
- load_definition raises FileNotFoundError instead of a generic exception
if the file was not found
- the string representation of several definitions is now
not so user friendly.
|
| |/
|
|
|
|
|
| |
Creating NaN-value quantities should take into account the
non_int_type, and create a magnitude of the correct type. Not all
common non_int_type choices support NaN (eg, Fraction does not); this
will likely result in a ValueError in that case.
|
| | |
|
| |
|
|
|
|
|
| |
registry
Prior to this commit, the facet registries could not be built correctly as the
dependent classes mro will not be properly ordered.
|
| | |
|
| |
|
|
|
|
| |
First we create the class when the registry class is subclassed
then we add a reference to the registry when the registry class
is instantiated.
|
| | |
|
| |
|
|
|
|
| |
Note: the word 'plain' was chosen instead of 'base' as BaseUnit or BaseQuantity
as the latter might be interpreted physically and not a part of the class hierarchy
as intented.
|
| |
|
|
| |
Still need to migrate basic implementation for Quantities and Units
|
| | |
|
| |\ |
|
| | |
| |
| |
| |
| |
| |
| | |
infer_base_unit accepted a UnitLike, but could not actually accept a
string or UnitsContainer. This uses to_units_container, and the
registry paramater, to try to accept all inputs, and throws a
ValueError if it can't.
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
to_compact used math.log10, which converted non-float types to floats;
this caused exceptions when using a non-integer type, like Decimal,
that does not implement multiplication/division with floats, and
generally changed non-integer types to floats. This addresses the
problem by checking non_int_type. If non_int_type != float, it
assumes that the type implements a log10 method, as Decimal does.
infer_base_unit created a generic UnitsContainer, not a UnitsContainer
specific to a registry; this allows specification of a registry, or
uses the registry of the passed-in Quantity/UnitLike.
|
| | |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
Starting from Python 3.3, the hash value of an object has a random salt
that is chosen at interpreter startup. Therefore, pickling and recovering
and object hash is incorrect if the the dumpeed value will be used at
a different interpreter run.
Until this commit, a UnitsContainer that was pickled and then unpickled
in a different interpreter would be wrongly consider as unequal.
BREAKING CHANGE: This commit will break existing code that relies on
keeping the hash across interpreters run. This should be very rare
as will require to set the PYTHONHASHSEED value.
|
| |/ |
|
| |\
| |
| | |
custom unit formats
|
| | | |
|
| |\ \
| |/ |
|
| | | |
|
| | | |
|
| |/ |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
| |
- Quantity as Generic class
- Add overloaded signature for __new__ Quantity
- Add typing module as private
- Add py.typed for PEP561 supports
- Add overloaded signature for __new__ Quantity
- Quantity as Generic class
- Add type hints throughout the project
- Add py.typed in package data in setup.cfg
- Add type hints for decorators
- Add type hints for public API of registry.py
- Add type hints for units.py
|
| | |
|