Tim/514 918 python data models

[linear[bot]]

514-918 python data models

[vercel[bot]]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
company-landing	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	May 2, 2024 11:20pm
framework-docs	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	May 2, 2024 11:20pm
framework-landing	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	May 2, 2024 11:20pm
moose-product-analytics	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	May 2, 2024 11:20pm

[phiSgr]

Since the types are retained in Python runtime (just ignored by the interpreter), maybe we should, instead of interring the semantics from the syntax, import the user's file then do typing.get_type_hints.

E.g.

>>> import simple
>>> dir(simple)
['Enum', 'Key', 'List', 'MyModel', 'Optional', 'Status', '__builtins__', '__cached__', '__doc__', '__file__', '__loader__', '__name__', '__package__', '__spec__', 'dataclass', 'field']
>>> type(getattr(simple, 'Status'))
<class 'enum.EnumType'>
>>> [element for element in getattr(simple, 'Status')]
[<Status.OK: 'ok'>, <Status.ERROR: 'error'>]
>>> type(getattr(simple, 'MyModel'))
<class 'type'>
>>> import typing
>>> typing.get_type_hints(getattr(simple, 'MyModel'))
{'name': Key[str], 'age': <class 'int'>, 'flag': <class 'bool'>, 'status': <enum 'Status'>, 'test_key': <class 'str'>, 'arr': typing.List[str], 'opt': typing.Optional[str]}
>>>

We can do sth very similar to how we read the data model config in ts.