« first day (2332 days earlier)      last day (848 days later) » 

12:00 AM
RELOAD!
[rubberduck-vba/Rubberduck] 2 issue comments.
[Minesweeper] Games Played: 103, Bombs Used: 70, Moves Performed: 14965, New Users: 9
 
 
1 hour later…
1:11 AM
@this No more work planned for the two I have in the queue.
 
 
2 hours later…
2:56 AM
[rubberduck-vba/Rubberduck] bclothier pushed 62 commits to next (only showing some of them below)
[rubberduck-vba/Rubberduck] BZngr pushed commit 7714e661 to next: Merge branch 'EF_IncorporateIndenter' into EF_CodeBuilderChanges
Expose IIndenter on ICodeBuilder

Also corrects merge error and minor changes to signatures and comments
Add ModifyUserDefinedTypeRefactoringAction

Replaces CreateUDTMemberRefactoringAction. Leverages ICodeBuilder and
IIndenter to rewrite the entire UDT declaration after adding (or
removing) UDT Members.
Rework EncapsulateFieldRefactoring factories

Let CW supply factories where possible. Resulted in removal of several
explicitly defined factory class files...which led to a lot of renames,
refactoring, and some file structure changes.
Make NoActiveSeletion_Throws() test virtual

Passing a null IDeclarationFinderProvider reference may cause some
subclasses to fail or throw an unexpected Exception prior to generating
the expected Exception. Support overriding when needed (e.g.,
EncapsulateFieldRefactoring tests)
Relocate to correct namespace

Move from RubberduckTests.Refactoring.EncapsulateField to
RubberduckTests
Introduce CW support for tests

Modified EncapsulateFieldRefactoring tests to use a WindsorContainer to
manage ctor injection and object instantiations - causing some degree of
modification to all EF related tests.
[rubberduck-vba/Rubberduck] BZngr pushed commit 84c2173f to next: Address PR comment
[rubberduck-vba/Rubberduck] BZngr pushed commit 9db29d95 to next: Add NewLines Resource class
Merge pull request #5578 from BZngr/EF_SplitRefactoring

EncapsulateFieldRefactoring - Split Refactoring
 
 
10 hours later…
1:53 PM
 
 
1 hour later…
3:04 PM
@this thanks! I've yet to get my VM server back up & running so the website (and thus version check) won't pick it up just yet, but what I'll do is I'll grab the code and last build from the mounted vhdx and run it manually
 
 
1 hour later…
4:20 PM
@MathieuGuindon but, why am I getting version check for last few days?
it even shows .5610 now
(unfortunately I wasn't paying attention yesterday or the day before when I got that prompt but I know I'm running a older version currently)
 
4:46 PM
Hm weird.. "Tag/release metadata updated 2020-09-29 05:09:38"
 
5:18 PM
I also do not plan to add anything more to my PRs unless there are review comments to address.
Regarding the possible conflicts with the German translation PR, I think that I will have to finalize that PR myself since I have not gotten any reaction so far.
 
5:30 PM
Do not issue neg line number results for On Error GoTo -1 unless there is a line number -1

This PR also fixes a declaration resolver bug that removed the minus sign from negative line labels.
[rubberduck-vba/Rubberduck] MDoerner pushed commit c5ec039a to next: Merge branch 'next' into FixNegLineNumberDetectedInGoToNegOne
Merge pull request #5594 from MDoerner/FixNegLineNumberDetectedInGoToNegOne

Fix neg line number detected in legitimate go to neg one
 
@M.Doerner note that 5587 is now conflicting and needs to be synchronized.
 
6:15 PM
> Closes #196 Closes #3225 This PR introduces a DeclareAsExplicitTypeQuickFix which calls an ImplicitTypeToExplicitRefactoringAction to evaluate implicit declarations for Variables, Constants, and Parameters. It could replace the DeclareAsExplicitVariantQuickFix, but this PR retains the existing DeclareAsExplicitVariantQuickFix as an option for the user. The RefactoringAction does not attempt to parse expressions. If a type inference opportunity encounters an expression,...
the RefactoringAction defaults to either Variant (emulating the current QuickFix) or Object. Once a unified expression evaluation capability is available (#3969), then the RefactoringAction can be enhanced to fold in expression evaluations as well. To evaluate literals, the IParseTreeValueFactory was recruited from the UnreachableCaseEvaluation inspection. This required moving ParseTreeFactory and a minimal set of supporting classes and interfaces to the...
Rubberduck.Refactorings project so that it could be available to both the QuickFixes/Inspectionsand Refactorings. #3969 discusses moving the ParseTreeFactory to another project (e.g., Rubberduck.Expressions) as part of creating a project dedicated to parsing expressions - but I did not want to attempt a start at that migration as part of this PR.
 
6:30 PM
@this Thanks for the heads up. I will update my PRs once I get home.
 
Fix missing default member result for optional parentheses on properties

The problem was that the empty argument list gets parsed as an argument list with a single missing argument and the IndexDefaultBinding did not account for it. Since the missing argument is actually a valid way to parse the empty argument list and the parser cannot know whether there is an optional parameter or not, I chose to catch this in the IndexDefaultBinding.
Merge pull request #5597 from MDoerner/FixDefaultMemberInspectionForOptionalPropertyParentheses

Fix missing default member result for optional parentheses on properties
[rubberduck-vba/Rubberduck] build for commit 5b7e35e3 on unknown branch: AppVeyor build succeeded
 
 
1 hour later…
7:57 PM
Your fame grows Mug. You’re called out by your peers: nolongerset.com/interfaces-practically
I used to think interfaces were “advanced really hard stuff”.
Strange how hard topics become mundane as you use them often enough.
 
@Duga Typical XML merge conflict
 
[rubberduck-vba/Rubberduck] build for commit a8c48730 on unknown branch: AppVeyor build succeeded
 
8:58 PM
[rubberduck-vba/Rubberduck] build for commit 6870b27e on unknown branch: AppVeyor build succeeded
 
 
1 hour later…
10:04 PM
Get document supertypes from user com projects

We use the user com projects to get the names of the implemented interfaces; the type hierarchy pass does the rest.

There is one hack in this: we suppress the underscore in front of the implemented interfaces in order to emulate the VBE's behaviour to let type checks pass relative to the coclass instead of the interface. This is entirely based on convention.

Moreover, this lacks unit test since the com projects cannot be mocked at this stage.
Change the document module supertype behaviour to adding supertypes without underscore

This seems safer in case we actually have a declaration for the interface with an underscore.
Fix NRE if the user com project for a document module does not exist

An appropriate guard clause was missing in the previous two commits.
Add ImplicitContainingWorkbookReferenceInspection and ImplicitContainingWorksheetReferenceInspection

These two find unqualified references inside the corresponding document modules to certain members that point to the ActiveWorkbook or ActiveSheet when used unqualified outside the appropriate document modules. That the reference is actually to the containing document can be surprising. Because of that, they should be qualified with Me.

This also removes the corresponding false-positives for
Adjust com interface discovery for documents with separate interface

Hosts like Access insert an additional interface for document modules between the document and the built-in one. Since we do not have a declaration for the separate interface, we also add the interface two levels op to the supertypes of the document.
Add QualifyWithMeQuickFix for ImplicitContainingWorkbook/WorksheetReferenceInspection

It does what the name suggests.
[rubberduck-vba/Rubberduck] MDoerner pushed commit 1e9d244a to next: Merge branch 'next' into SuperTypesForDocumentModules
Stop adding super type names for documents that are already there

This fixes a small memory leak in the reference resolver. Although the supertypes of a class declaration are a hash set, the super type names are a list. So, adding all supertype names derived from the typeLib API every parse would grow that list unnecessarily.
Merge branch 'next' into SuperTypesForDocumentModules

# Conflicts:
# Rubberduck.Resources/Inspections/InspectionInfo.resx
# Rubberduck.Resources/Inspections/InspectionNames.resx
# Rubberduck.Resources/Inspections/InspectionResults.resx
Merge pull request #5587 from MDoerner/SuperTypesForDocumentModules

Super types for document modules
 
11:10 PM
> I resolved the typical XML conflict and snug in a fix for the other type on the same line as the original one.
[rubberduck-vba/Rubberduck] build for commit 9e54fb6c on unknown branch: AppVeyor build succeeded
 
11:50 PM
[rubberduck-vba/Rubberduck] build for commit a28d7ed7 on unknown branch: AppVeyor build succeeded
 

« first day (2332 days earlier)      last day (848 days later) »