Validate if locals/args exist when adding a type declaration #603
Conversation
|
there seems to be a broken part of the build because of the PR :) |
| { #category : #translation } | ||
| SLBasicTestDeclarationClass class >> typeForSelf [ | ||
|
|
||
| ^#implicit | ||
| ] |
There was a problem hiding this comment.
Why is this necessary? Is it not always implicit (except for structs)?
There was a problem hiding this comment.
The inherited behaviour from SlangClass is to return nil. Or are you saying that at some point that nil defaults to #implicit?
There was a problem hiding this comment.
I was just asking for understanding, because this definition is in the tests, but maybe could be in a more general place? I don't know.
Anyway, it is out of the scope of this PR, so we can continue with our life.
| @@ -179,7 +179,6 @@ ComposedImageReader >> readPermanentSpaceDataFromImage: imageFileName startingAt | |||
| ComposedImageReader >> readPermanentSpaceFromImageFile: imageFileName header: aHeader [ | |||
|
|
|||
| <inline: false> | |||
| <var: #imageFile type: #sqImageFile> | |||
|
Failing tests are not related. Windows is not compiling, but neither is in the main branch. I think that it is mergeable 馃憤 |
|
I just fixed a merge conflict, I'll wait the CI to see if I did well :) |
|
Tests are green!! Thanks!! |
fix #598
Adds a validation at
TMethod >> declarationAt: aVariableName put: aDeclarationRemoved unused type declarations and then generated the VM C Code and compiled succesfully 馃殌