« first day (1198 days earlier)      last day (1982 days later) » 

Duga
Duga
00:00
RELOAD!
[rubberduck-vba/Rubberduck] 1 opened issue. 2 issue comments.
IvenBach
IvenBach
Home Time!
Duga
Duga
00:16
[rubberduck-vba/Rubberduck] ThunderFrame commented on issue #3375: Bitwise annotation for Enums, and Flags property for declarations
> @bclothier Well, Microsoft [marks](https://referencesource.microsoft.com/#Microsoft.VisualBasic/Microsoft/VisualBasic/MsgBoxStyle.cs) it with a `Flags` attribute, so presumably the 0-5 values are designed to be mutually exclusive, much like the `VbStrConv.ProperCase` example I gave.

When it's a COM-defined enum, that RD sets as being Flags, RD won't be offering any inspections on the Enum itself, but rather on usages of the enum that, for example, use `+` instead of `Or`.

When it's a us
Hosch250
Hosch250
00:46
Looks much better, @IvenBach.
 
1 hour later…
Duga
Duga
02:00
[rubberduck-vba/Rubberduck] IvenBach synchronized pull request #3376: Consolidate Empty(If|Else)Block into EmptyConditionBlock
[rubberduck-vba/Rubberduck] IvenBach commented on Rubberduck.Inspections/Concrete/EmptyConditionBlockInspection.cs of pull request #3376: Consolidate Empty(If|Else)Block into EmptyConditionBlock
IvenBach
IvenBach
@Hosch250 What looks better?
Duga
Duga
[rubberduck-vba/Rubberduck] build for commit b44850b5 on unknown branch: AppVeyor build succeeded
 
1 hour later…
IvenBach
IvenBach
03:27
So my daughter challenged me with drawing something awesome.
4
user image
 
7 hours later…
Duga
Duga
10:08
[rubberduck-vba/Rubberduck] rkapka commented on Rubberduck.Inspections/Concrete/ShadowedDeclarationInspection.cs of pull request #3252: ShadowedDeclarationInspection
[rubberduck-vba/Rubberduck] rkapka commented on Rubberduck.Inspections/Concrete/ShadowedDeclarationInspection.cs of pull request #3252: ShadowedDeclarationInspection
Duga
Duga
10:20
[rubberduck-vba/Rubberduck] rkapka commented on Rubberduck.Inspections/Concrete/ShadowedDeclarationInspection.cs of pull request #3252: ShadowedDeclarationInspection
Duga
Duga
10:53
[rubberduck-vba/Rubberduck] bclothier commented on Rubberduck.Inspections/Concrete/ShadowedDeclarationInspection.cs of pull request #3252: ShadowedDeclarationInspection
 
1 hour later…
Duga
Duga
12:09
[rubberduck-vba/Rubberduck] rkapka commented on Rubberduck.Inspections/Concrete/ShadowedDeclarationInspection.cs of pull request #3252: ShadowedDeclarationInspection
Duga
Duga
12:54
[rubberduck-vba/Rubberduck] retailcoder merged pull request #3252: ShadowedDeclarationInspection from rkapka/Rubberduck/rkapka-master into rubberduck-vba/Rubberduck/next
[rubberduck-vba/Rubberduck] retailcoder pushed 26 commits to next (only showing some of them below)
this
this
Looks to me that an ExitStmtContext can represent any exits, including Exit Do, Exit For and as well Exit <procedure>. What is the best way to determine if the context is actually the Exit <procedure> variety, excluding Exit Do/Exit For?
Duga
Duga
[rubberduck-vba/Rubberduck] retailcoder published prerelease Rubberduck-v2.1.0.2224
[rubberduck-vba/Rubberduck] retailcoder created tag Rubberduck-v2.1.0.2224
[rubberduck-vba/Rubberduck] build for commit 4a827925 on next: AppVeyor build succeeded
Mat's Mug
Mat's Mug
13:10
@this looking at its tokens I guess, no?
Also, #DidYouKnow you can Exit Sub and Exit Function in a Property member?
this
this
yeah, i think i can look at the token - just wanted to see what was the best way. I'll do that.
Yes i knew but thanks for reminding me about that weird thing
in my specific case, it should not apply since I really don't care what procedure it is; I just simply want no Exit <procedure> in the selection
the Exit Do/ Exit For would require analysis, to verify that the selection contains the outer block.
Mat's Mug
Mat's Mug
Man that refactoring is the funnest one to implement - wait till you get an actual procedure extracted, that moment of utter pride is priceless =)
this
this
just so I know -
if I need to add strings for messaging
I'm supposed to use the .resx files
Mat's Mug
Mat's Mug
Yup
this
this
but what would happen if german/french didn't have the correspodning stuff?
I mean, I can use Google Translate but.... ? (would hate it to come out sounding dirty or at least moronic)
Mat's Mug
Mat's Mug
13:24
The framework defaults to en-US for the missing keys
this
this
good - and i hope it's easy to find what are missing whenever someone else does localization work later on?
Mat's Mug
Mat's Mug
i.e. don't bother with them, they'll be handled =)
Yep, @Vogel612 wrote a tool specifically for that
this
this
great!
Mat's Mug
Mat's Mug
kids' swimming lessons, gotta go
later!
 
1 hour later…
Duga
Duga
14:55
[rubberduck-vba/Rubberduck] retailcoder commented on RubberduckTests/QuickFixes/RemoveEmptyConditionBlockQuickFixTests.cs of pull request #3376: Consolidate Empty(If|Else)Block into EmptyConditionBlock
[rubberduck-vba/Rubberduck] Hosch250 commented on Rubberduck.Inspections/Concrete/EmptyConditionBlockInspection.cs of pull request #3376: Consolidate Empty(If|Else)Block into EmptyConditionBlock
Hosch250
Hosch250
@IvenBach This new implementation vs. the old one.
M.Doerner
M.Doerner
15:48
Everything is working now with Castle Windsor, but the access violations on exit remain as well as the crash on exit with the native debugger attached.
I also tested the startup time and Castle Windsor seems to be slightly faster, but that might as well be a difference in system load. (Depensing on what is running on my system, the load time can vary by 1s.)
IvenBach
IvenBach
16:45
@Hosch250 I'm hoping to get time to improve it more. I used @this's suggestion and it looks better but still want to see if it will tidy up more.
 
1 hour later…
this
this
17:48
@M.Doerner going out on the limb but doesn't it imply that there might be a wrong lifecycle used for a given VBE component that's injected? IOW, there might be something managed as a singleton when it should be transient (or vice versa, even)
 
2 hours later…
M.Doerner
M.Doerner
19:37
I have my doubts that this has something to do with wrong lifestyles because the access violations start after the container has been disposed.
I rather think that something not managed by the container does not get unloaded properly.
Hosch250
Hosch250
20:22
Iirc, transient lifecycle items aren't disposed by the system.
M.Doerner
M.Doerner
In Castle Windsor, they are.
Hosch250
Hosch250
20:51
Cool. That was always pain in Ninject.
this
this
@M.Doerner you know, it might be a good idea to try and port the Castle Windsor to the SlimDucky project to also prove that it's not something inherent to CW but is in fact our fault.
For all I know, it might be that the whole DI paradigm might be anathema toward COM reference counting. IDK.
(Mind you, I sincerely I"m totally wrong on this)
 
2 hours later…
ThunderFrame
ThunderFrame
23:11
Just thinking about adding declarations for WordBasic. For example, SelType can be used as a statement Application.WordBasic.SelType 1 (to set the Selection type), or as a function, Debug.Print Application.WordBasic.SelType() (to return the current Selection type). The statement form accepts an integer parameter value between 1 and 7. The return value is an integer value of 1,2,5 or 6.
I can add SelType as a FunctionDeclaration. For the statement form, do I add it as a SubroutineDeclaration?
And I can define the parameter type and return type as Integer, but I don't think there is a way of defining the ranges of valid values?

« first day (1198 days earlier)      last day (1982 days later) »