Peter Law
ea33db388b
Remove dict merging where it doesn't do anything
...
These cases are all at the end of a single-path branch that ends
up "merging" against an empty mapping which is then returned
unchanged.
2020-03-22 15:45:18 +00:00
Peter Law
f68d65ed59
Push much looping and merging of infering type vars into ValueSet
2020-03-22 15:29:11 +00:00
Peter Law
3c7621049c
Extract annotation inference onto annotation classes
...
This removes the _infer_type_vars util in favour of a polymorphic
implementation, removing the conditional checks on the type of
the annotation instance.
While for the moment this creates some circular imports, further
refactoring to follow should be able to remove those.
2020-03-22 15:29:11 +00:00
Peter Law
dd60a8a4c9
Extract nested function which is going to be used elsewhere
2020-03-22 15:20:58 +00:00
Peter Law
5bd6a9c164
Rename function which is going to be used elsewhere
2020-03-22 15:18:41 +00:00
Peter Law
c743e5d9f3
Push type check into helper
2020-03-22 15:14:01 +00:00
Peter Law
5ca69458d4
Add testing for mismatch cases
...
This should help catch any errors in our handling of invalid cases.
While some of these produce outputs which aren't correct, what
we're checking here is that we don't _error_ while producing that
output.
Also fix a case which this showed up.
2020-03-22 15:10:43 +00:00
Peter Law
0f8e7b453e
Formatting
2020-03-18 22:12:21 +00:00
Peter Law
da9d312185
Remove redundant attribute check
2020-03-12 22:06:13 +00:00
Peter Law
b198434694
Remove resolved TODO
...
The common logic this refers to has now been extracted (see 95cec459 )
and the remaining checks are specific to tuple handling.
2020-03-07 20:29:14 +00:00
Peter Law
d06efd0dd1
Push fetching of generics into nested function
...
This slightly simplifies both the calling code and semantics of
the nested function.
2020-03-07 18:09:20 +00:00
Peter Law
96132587b7
Clarify generic tuple inference
...
This hoist a loop invariant conditional check outside the loop
making it clearer and one branch more obviously similar to the
general type handling.
2020-03-07 17:35:29 +00:00
Peter Law
5d273f4630
Explain these branches
2020-03-07 17:35:03 +00:00
Peter Law
95cec459a8
Extract nested function for common pattern
...
This slightly simplifies the code, as well as providing a place
to put an explanation of what the moved block of code does.
2020-03-07 17:06:22 +00:00
Peter Law
3b4fa2aa9c
Clarify variable name
2020-03-07 16:32:38 +00:00
Peter Law
54e29eede1
Add explanation of the parameters to _infer_type_vars
2020-03-07 16:31:12 +00:00
Peter Law
f1a9e681ad
Ensure comprehensions and generator expressions work
2020-02-23 15:25:28 +00:00
Peter Law
f4cbf61604
Ensure variadic tuples (Tuple[T, ...]) behave like sequences
2020-02-23 14:00:39 +00:00
Peter Law
e557129121
Remove check which doesn't seem to be needed
...
I'm not sure why I added this, though removing it doesn't seem to
casue any issues. I suspect there might be some oddness if the type
being passed in doesn't match the type expected, though them having
the same number of generic paramters isn't an expecially great way
to validate that.
2020-02-23 14:00:16 +00:00
Peter Law
c15e0ef9b8
Ensure specialised types inheriting from generics work
2020-02-23 14:00:15 +00:00
Peter Law
c03ae0315e
Make nested Type[T] annotations work
2020-02-23 13:59:44 +00:00
Peter Law
bc53dabce3
Make tuple generic parameters work
2020-02-23 13:59:44 +00:00
Peter Law
969a8f1fd9
First pass at extending infer_type_vars
...
This mostly works for the new tests, but doesn't work for:
- tuples (though this seems to be because they lack generic information anyway)
- nested Type[T] handling (e.g: List[Type[T]])
2020-02-23 13:59:44 +00:00
Peter Law
6efafb348e
Extract the annotation name upfront
...
We almost always need this and this simplifies the code within
each branch. This also means we'll be able to the name to determine
the branching.
2020-02-22 19:42:08 +00:00
Peter Law
36b4b797c1
Add trailing comma
2020-02-22 19:42:08 +00:00
Dave Halter
7a55484b79
Fix a test issue
2020-02-04 23:56:01 +01:00
Dave Halter
a49c757b8a
Make Ellipsis without list in Callable work, fixes #1475
2020-02-03 09:25:46 +01:00
Dave Halter
eee919174d
Stubs should not become stubs again in the conversion function, fixes #1475
2020-02-03 00:58:54 +01:00
Dave Halter
e3c4b5b77e
Make sure param hints are working for functions
2020-02-02 18:42:01 +01:00
Dave Halter
4c7179bc87
Generate type hints, fixes #987
2020-02-02 16:55:10 +01:00
Dave Halter
5c68304bec
Raise a proper exception instead of assert in case only_stubs and prefer_stubs are given
2020-01-22 10:00:10 +01:00
Dave Halter
8bcd1f5fd9
Fix stub conversion
2020-01-14 01:08:26 +01:00
Dave Halter
095f1295af
Avoid a bug that a compiler might have found, fixes #1469
2020-01-11 21:35:39 +01:00
Dave Halter
3ba68b5bc6
Properly convert compiled values to generic classes
2020-01-10 15:09:16 +01:00
Dave Halter
a17b56f260
Use one single way to convert stubs to Python, see #1466
2020-01-07 10:02:31 +01:00
Dave Halter
9b9cacfbf9
Make sure to use _stub_to_python_value_set for all conversions, see #1466
2020-01-07 01:27:50 +01:00
Dave Halter
ea0972d7ac
Make sure to check the module cache before loading a module (again)
...
This hopefully results in some performance improvements (maybe numpy?).
2020-01-05 18:28:34 +01:00
Dave Halter
829ee0e6b0
Remove unused code
2020-01-01 03:27:17 +01:00
Dave Halter
8611fcf8ea
Fix some tests
2019-12-27 11:59:40 +01:00
Dave Halter
4161bfc7f2
Avoid some duplication of code
2019-12-22 01:24:50 +01:00
Dave Halter
fcede44c2a
Move the docstring checking code to the names
2019-12-21 20:06:37 +01:00
Dave Halter
e56d4fde98
Improved Generic subclass matching
2019-12-20 01:33:41 +01:00
Dave Halter
51e2e90dce
Make sure overload checks work for TypeAlias, see #1416
2019-12-20 01:33:41 +01:00
Dave Halter
38eb2c9ba3
Make sure that the definition order in stubs are ignored
2019-12-15 18:41:41 +01:00
Dave Halter
9d35adda02
Make sure that a goto on stubs even without the implementation is possible
2019-12-15 18:00:09 +01:00
Dave Halter
7bdedb40e3
Fix: Stubs in typeshed weren't loaded properly sometimes, fixes #1434
2019-12-14 21:33:00 +01:00
Dave Halter
e656a5f18f
Make it possible to infer Callable TypeVars, fixes #1449
2019-12-12 23:46:55 +01:00
Dave Halter
536a77551b
Account for sys path potentially not being all unicode in typeshed, fixes #1456
...
This is a bit stupid, but don't care too much, it will get removed soon, when
Python 2 support is going to get removed.
2019-12-11 00:30:37 +01:00
Dave Halter
28ecc2709a
Don't use globals anymore
2019-12-09 00:15:21 +01:00
Dave Halter
33224ae7e1
Remove a duplicate method
2019-12-09 00:11:51 +01:00