« first day (1115 days earlier)      last day (2065 days later) » 

12:00 AM
RELOAD!
[Hosch250/Rubberduck] 116 commits. 48474 additions. 27871 deletions.
[rubberduck-vba/Rubberduck] 4 commits. 3 issue comments. 92 additions. 4 deletions.
 
12:24 AM
> Rubberduck.Setup.2.0.13.0.exe (5.83 MiB) - Downloaded 2171 times.
Last updated on 2017-03-12
> Total Downloads 12,466
2
 
 
11 hours later…
11:26 AM
@Mat'sMug question: when will the next release go out?
IMO the next release should go out somewhen soonish ...
2
drop me a note when you set a date, because then I'd fix up the German translations ...
 
 
4 hours later…
3:19 PM
@Vogel612 agreed. I was hoping for early July, but I'm not sure how feasible that would be; there's still a thing or two to fix up with the attributes-rewriting quickfix (and it doesn't do member attributes yet)
 
Anything required for the next release?
BTW, my brother is learning programming.
Maybe in a year or two I'll introduce him to Rubberduck.
Depends on how fast he is at learning.
 
[rubberduck-vba/Rubberduck] Hosch250 pushed commit d683db06 to next: Fix quick fix provider bug that only rewrites one module
[rubberduck-vba/Rubberduck] build for commit dbebaba5 on next: AppVeyor build succeeded
 
4:07 PM
@Mat'sMug I'm totally swamped until July 2nd... after that I'd be ready for release at a days notice or something
 
 
2 hours later…
6:08 PM
@Hosch250 yes: the rest of the SynchronizeModuleAttributesQuickFix, and a way to reopen/activate a code pane that was open/active before the quickfix exported & reimported the module.
 
Don't we have that already?
Look at the Find All References thing--we open code panes...
 
not that I know of
the COM object isn't the same before & after importing. Totally different case here.
 
Oh.
I don't think that's an issue. The name is the same.
We can do vbe.projects[name].components[name].codemodule.
 
Well if you try to call a method on the stale COM object you're in for a nice little COMException
Probably. I just can't even think of turning on my computer ATM
 
You don't need to use the stale com object.
I just told you how to get the fresh one.
 
6:21 PM
I don't know when I'll be able to sit for a whole evening at my computer. Migt be some time next week, might be the week after. You're in charge @Hosch250 =)
 
Not sure I want to be.
 
Basically we want to resolve the conflicts on the 2 pending PRs, get them merged, and then feature-freeze (e.g. no new inspections) while we finish and polish what's there.
 
Ivan's doesn't have a quick fix, which will look stupid since the others do. Also, this should be merged with my existing one for DRY, SRP, etc., and the whole thing should be renamed.
The other one has a bug (see my comment).
Neither of these has serious merge conflicts (just the .resx files and the .csproj for the tests), but they don't allow maintainers to edit.
 
So the two of them are basically stuck then
 
Something.
 
7:09 PM
 
 
2 hours later…
9:08 PM
[rubberduck-vba/Rubberduck] build for commit 7a82a439 on unknown branch: AppVeyor build failed
BUILD FAILURE!
 
10:06 PM
Hi
 
[rubberduck-vba/Rubberduck] build for commit 798ab456 on unknown branch: AppVeyor build succeeded
> I noticed that parameter names in interface declaration can differ from the names in implementation declarations. Because of this I am relying on the position of the parameter, although I don't know if `IReadOnlyList` returned from `VBAParser` will give me the right order every time.

Also, it is possible to explicitly declare `ByRef` in the implementation, but not in the interface (and the other way around, but that's not a problem). Such a combination won't be listed as an inspection. Do w
 
10:33 PM
> Actually, ByRef in the implementation/not in the interface and the other way around, is a problem. The user might not set out to write code like that, but it's easy to envisage a situation where they refactor one without refactoring the other(s). Assigning to a variable that you think is ByRef, but is actually ByVal can crash the host - see #3054
 
10:58 PM
Hi, @Kapol.
We were just discussing syncing that PR.
You are right that you need to rely on position; I learned this when working on Reorder/Remove parameters.
I don't think it is an implementation guarantee that a list will return the results in the right order (I think Skeet actually warns about people relying on this), but it usually does, so...
If you are iterating the list from the grammar, I think you are probably pretty safe. You could get the list of parameters and order by position if you really want--the performance cut will be unnoticeable.
 

« first day (1115 days earlier)      last day (2065 days later) »