« first day (1423 days earlier)      last day (1757 days later) » 

Duga
Duga
12:01 AM
RELOAD!
[bruglesco/memory-card] 1 commit. 101 additions. 21 deletions.
[Cardshifter/HTML-Client] 6 commits. 131 additions. 324 deletions.
[rubberduck-vba/Rubberduck] 1 opened issue. 5 issue comments.
 
 
7 hours later…
QHarr
QHarr
7:01 AM
morning
 
 
8 hours later…
Duga
Duga
2:45 PM
 
IvenBach
IvenBach
3:19 PM
@Vogel612 I should get time today to see your PR for the about window. Going out of my comfort zone again to do a pr into my repo
 
Duga
Duga
3:47 PM
 
Mathieu Guindon
Mathieu Guindon
4:05 PM
@Duga @this does your analyzer pick these up?
(that would be insanely awesome)
 
this
this
It could. we just need to write that code
which is why I went to the trouble of setting the project up in first place. It's precisely because of that, which is by far most our painful aspect of using the wrappers.
Maybe Hosch can write it out quicker than I could?
 
Mathieu Guindon
Mathieu Guindon
@Hosch250 you're up! ^^
 
this
this
Right now the analyzer I wrote will inspect for COM visible types and enforce 1) Guid attribute on all, 2) InterfaceType on the interface types, 3) ProgId and ClassInterface on class types.
 
Mathieu Guindon
Mathieu Guindon
You haven't PRd it yet though right? Or did I miss it?
 
this
this
it additionally will disallow literals in Guid and ProgId, insisting that it be Guid(RubberduckGuid.... and ProgId(RubberduckProgId...
No it's still in my working branch
It really should have been its own PR but I painted myself into the corner with the API cleanup
 
Mathieu Guindon
Mathieu Guindon
4:08 PM
@this that's awesome, RD rubberducking itself
2
 
this
this
Once I'm finished cleaning up the API (note lot lot of breaking changes) it'll be PR'd
Yep! I decided that wiki article wasn't going to cut it.
Since everyone was basically ignoring @comintern's guideline from last year. :)
 
 
1 hour later…
Duga
Duga
5:10 PM
> # [Codecov](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/3973?src=pr&el=h1) Report
> Merging [#3973](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/3973?src=pr&el=desc) into [next](https://codecov.io/gh/rubberduck-vba/Rubberduck/commit/1b1fe5cb0a9f3f17e4016128b4808fda16af7982?src=pr&el=desc) will **increase** coverage by `0.01%`.
> The diff coverage is `100%`.


```diff
@@ Coverage Diff @@
## next #3973 +/- ##
=========================
 
Hosch250
Hosch250
5:31 PM
@MathieuGuindon What's going on?
@this wrote a Roslyn analyzer?
 
this
this
yeah I did
it's meant to be solution-specific, though.
 
Hosch250
Hosch250
Yep.
 
this
this
anyway - the chained member access is very easy thing for us to miss
so if we can add a new rule to call out those leaks, that would be totally awesome.
 
Hosch250
Hosch250
Well, what I'd do is listen for chained calls, then get the type for the expression.
Think you need to use the SemanticModel or something (forget the name).
 
Duga
Duga
> If a declaration statement has a line number preceding it, the line number will be left behind when moving the declaration. This probably makes sense because the number may not be in sequence if it was moved too (although that seems to be valid VBA). However, since the following line is moved up, that line inherits the line number and if it already has one, we get a line with two numbers as per the following example:

Before refactoring:
```vb
Sub SimpleLineNumber()
1 Dim a As Long
2
[rubberduck-vba/Rubberduck] build for commit 22181b00 on unknown branch: AppVeyor build succeeded
 
Hosch250
Hosch250
5:36 PM
Has Tommy been invited here?
 
this
this
I see. The OM for the analyzer is quite different from the OM that we use for RD's parsing, and to be honest, I cowboyed it a little w/ the analyzer because the data I needed weren't in the places where I expected to be.
 
Hosch250
Hosch250
Yeah, ping me when you PR it.
 
this
this
Deal
 
Duga
Duga
> Good job. Feel free to join us in our war room: chat.stackexchange.com/rooms/14929/vba-rubberducking
[rubberduck-vba/Rubberduck] tommy9 pushed commit 4edaa1e8 to next: Preserve indentation of expression following inserted declaration (and associated tests)
[rubberduck-vba/Rubberduck] tommy9 pushed commit a556bb49 to next: Request reparse after move closer to usage refactoring
[rubberduck-vba/Rubberduck] tommy9 pushed commit 54de2551 to next: Wrap ActiveCodePane and CodeModule in using blocks
[rubberduck-vba/Rubberduck] tommy9 pushed commit 22181b00 to next: Test case for MoveCloserToUsage refactoring with line numbers
Merge pull request #3973 from tommy9/MoveCloserIndentFix

Move closer refactoring indent fix
 
Duga
Duga
5:52 PM
[rubberduck-vba/Rubberduck] build for commit a154ce8e on next: AppVeyor build succeeded
 

« first day (1423 days earlier)      last day (1757 days later) »