Fix checking multiple assignments based on tuple unpacking involving partially initialised variables (Fixes #12915). #14423
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…partially initialised variables (Fixes python#12915). This commit introduces two changes: * It converts unions of tuples to tuples of unions during multi-assignment checking when all (original) tuples have the same length. This fixes issue python#12915. * It removes the `undefined_rvalue` logic completely. This logic bothered me because it introduced the necessity to make the mentioned conversion twice (which was not always successful due to reasons I do not fully understand). Test new case `testDefinePartiallyInitialisedVariableDuringTupleUnpacking` covers issue python#12915. I had to adjust some other test cases. In my opinion, the error messages are now more transparent, consistent, and complete. Removing the `undefined_rvalue` logic also seems to fix one more bug. In the test cases `testInferenceWithTypeVariableTwiceInReturnType` and `testInferenceWithTypeVariableTwiceInReturnTypeAndMultipleVariables`, there were three assignments for which Mypy did not report errors but, in my opinion, should. So, for the current test data set, removing the `undefined_rvalue` logic seems to introduce mere benefits than drawbacks (meaning, different error messages than before). I hope it does not break any Mypy functionalities not covered by the tests.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
This comment has been minimized.
tyralla
added a commit
to tyralla/mypy
that referenced
this pull request
Jan 12, 2023
…partially initialised variables (Fixes python#12915). This proposal is an alternative to python#14423. Similar to python#14423, the main idea is to convert unions of tuples to tuples of (simplified) unions during multi-assignment checks. In addition, it extends this idea to other iterable types, which allows removing the `undefined_rvalue` logic and the `no_partial_types` logic. Hence, the problem reported in python#12915 with partially initialised variables should be fixed for unions that combine, for example, tuples and lists, as well. Besides the new test case also provided by python#14423 (`testDefinePartiallyInitialisedVariableDuringTupleUnpacking`), this commit also adds the test cases `testUnionUnpackingIncludingListPackingSameItemTypes`, `testUnionUnpackingIncludingListPackingDifferentItemTypes`, and `testUnionUnpackingIncludingListPackingForVariousItemTypes`.
|
According to mypy_primer, this change has no effect on the checked open source code. 🤖🎉 |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fix checking multiple assignments based on tuple unpacking involving partially initialised variables (Fixes #12915).
This commit introduces two changes:
undefined_rvaluelogic completely. This logic bothered me because it introduced the necessity to make the mentioned conversion twice (which was not always successful due to reasons I do not fully understand).Test new case
testDefinePartiallyInitialisedVariableDuringTupleUnpackingcovers issue #12915.I had to adjust some other test cases. In my opinion, the error messages are now more transparent, consistent, and complete. Removing the
undefined_rvaluelogic also seems to fix one more bug. In the test casestestInferenceWithTypeVariableTwiceInReturnTypeandtestInferenceWithTypeVariableTwiceInReturnTypeAndMultipleVariables, there were three assignments for which Mypy did not report errors but, in my opinion, should.So, for the current test data set, removing the
undefined_rvaluelogic seems to introduce mere benefits than drawbacks (meaning, different error messages than before). I hope it does not break any Mypy functionalities not covered by the tests.