« first day (3138 days earlier)      last day (42 days later) » 

00:08
REFRESH!
[Minesweeper] 97 Games Played. 44 Bombs Used. 12480 Moves Performed. 20 New Users
[rubberduck-vba/Rubberduck] 10549 additions. 12528 deletions. 21 commits. 13 issues closed. 8 issue comments
[Rubberduck] 2 Synchronizations
00:55
@MathieuGuindon that's a huge number of closed issues :)
love to see it
01:26
> Because I don't have an ARM64 processor and because there seems to be an issue with using the build to build an ARM output, this makes things a bit more complicated so before we go to the trouble, I want to confirm this is a working version of ARM64 build. If any of you can confirm that this is working, we can look into integrating this in the build process.

This is the setup file as a zip since GitHub doesn't allow uploading exe file directly. Unzip it then run the setup. This will not run
 
2 hours later…
03:25
> hey @bclothier thank for this. I have attempted to get this working , but the COM addin does not seem to want to load. The installer runs fine , but I get "rubberduck add-in failed to load" error when launching the IDE.
I have tried installing for single user as well as all users. I was able to run the .bat file sucessfully but no change in the error message .
I also tried to run the regasm command from the the troubleshooting guide , and it resulted in a "rubberduck.dll is not a valid
> This feature is happening in v3, but isn't going to work (not nicely anyway) with the 2.x structure. Deferred to v3, since v3 is where Rubberduck gets to fully control the editor (by making our own!); automatic (using an idle timer reset on keypress) parsing proof-of-concept passed with flying colors - with parsing actually happening asynchronously and in a separate process in v3 (that part is WIP), consider it done 😉
 
1 hour later…
04:46
> hey @bclothier thanks for the zip.
Same thing here. I installed it as administrator, without problem. Then got the following error message when opening VBA or going to the add-in manager :

<img width="330" alt="image" src="https://user-images.githubusercontent.com/65823953/211180783-ed50747a-9063-4428-9eae-af9ff982c8d9.png">

on Microsoft® Excel® for Microsoft 365 MSO (Version 2211 Build 16.0.15831.20202) 64-bit
> I did some research and one thing I noticed is that the Rubberduck's DLL files are reported as ARM64 so that's right. I determined this using [ROM Properties](https://github.com/GerbilSoft/rom-properties). However, the dependent DLLs that aren't owned by Rubberduck does not appear to be generated as ARM64. I do not know if that's a factor but if the dependent DLLs fail to load due to incompatibility, that might be why.

Furthermore, Rubberduck depends on [Easyhook](https://github.com/EasyHoo
 
1 hour later…
06:08
> Hi guys! I am just getting started with using Visual Studio just for the purpose to help you guys somehow and to getting involved in the project as a thanks message for all your hard work.

I just added a few Spanish translations (which is my native language) just to make sure I am not messing anything up. If you guys can verify that I did the right thing. Then, I have the green light to put more work into it and create other pull requests later.

Please let me know if I need to do somethi
06:38
[rubberduck-vba/Rubberduck] build for commit 0dab7c4a on unknown branch: AppVeyor build succeeded
07:09
> Hy @bclothier, I also installed the Rubberduck ARM64 setup, so far have the same issue.
Don't know any details maybe this screenshot tells you more to te problem:

![image](https://user-images.githubusercontent.com/25910991/211184604-45c43787-e670-4415-8d40-d1aaf9bc8de1.png)

It displays as loaded but missing extension ... ?

Microsoft® Excel® 2016 MSO (16.0.14332.20431) 64-bit

Thanks & Cheers
Zoltan.
> Hy @bclothier, I also installed the Rubberduck ARM64 setup, so far have the same issue.
Don't know any details maybe this screenshot tells you more to te problem:

![image](https://user-images.githubusercontent.com/25910991/211184604-45c43787-e670-4415-8d40-d1aaf9bc8de1.png)

It displays as loaded but missing extension ... ?

*Edit:
Also checked the VBIDE's Tools -> References -> Rubberduck.x64.tlb

![image](https://user-images.githubusercontent.com/25910991/211184794-daeb5054-957a-4
 
8 hours later…
15:20
> Ok, so I got the confirmation that yes, `EasyHook` indeed can only run on x86 architecture and we would need to get an ARM build of `EasyHook` before we can then run Rubberduck on ARM. In the light of that information, we're effectively blocked on this until `EasyHook` can provide an ARM build.

Incidentally, it should be pointed out that Rubberduck has a number of other dependencies as well. For example, the [Castle Windsor](https://github.com/castleproject/Windsor) makes no mention of any
16:46
> @bclothier Thanks for efforts , unfortunate that external dependencies are blocking this.

A quick look at the easyhook repo shows that this has been discussed since the ARM based WinRT came out (7 yrs ago!) and there was never a resolution.

Maybe with the introduction of the M series hardware there will be more interest in ARM compatibility.

I have Win64 resources to run things in case , but have been trying to reduce my machine sprawl and have been moving to the MBP with parallels for Wi
17:01
> Parallels is a virtualization software rather than an emulator so if your MBP is based on an ARM processor rather than an Intel processor, then Parallels has to work with ARM-based Windows software. An emulator would be much slower. If you have any Mac that still has an Intel processor, then Parallels should be able to allow you to use x86-compiled software. I know that really doesn't help with the machine sprawl, though.
17:49
> Thanks @bclothier for enlightment.

Have some of you found Rubberduck replacement solution, by any chance, running on ARM ? Even paying product ?
@Duga would tB work on ARM64?
18:05
not right now but Wayne does have a plan to use LLVM, which should then make it possible to compile it for ARM64
> IIRC EasyHook is only used for hooking the Fakes API (and the "beep" suppressor for autocompletion features); if it's the only thing holding back an ARM64 build, there's probably a way to fork RD and build it without this dependency (obviously at the cost of the Fakes API not making it). As for a replacement, I'm not sure about ARM64 but I think twinBASIC is a serious contender.
@Duga You're forgetting the Castle. It's still not clear to me whether it's capable of hooking into other .NET assemblies compiled for Any CPU but the impression is that it has to be ARM64 all the way down to the last dependency. The fact that I can't get the Microsoft's dlls targeting ARM64 to go into the output doesn't make much sense. In EasyHook (as well as olewoo but that's not a runtime dependency), there are native dependencies.
Also, I thought we were using EasyHook for keyboard interception from VBEditor?
tbh I don't quite get the whole idea of providing an arm64 build. I mean sure let's supply an installer and make it as easy as possible to use RD, ..but it remains OSS, someone that wants something that the maintainers can't do, should fork the repo and do it themselves.
@this yeah we're suppressing the beeps in autocompletion
No, referring to the shortcuts. I recall we're hooking into keyboard to enable keyboard shortcuts but not sure if htat's using EasyHook
@Duga bummer - race condition still existsz.
Hm I don't recall how the hotkeys work. Wasn't it a low-level key hook? Or maybe it was and then it changed
RD3 hotkeys should be much simpler, since focus won't constantly be jumping in and out of WPF
Yes, hopefully. The hacking of IDE is a huge mudball
Don't we actually listen for keyboard window messages in the toolwindows to capture the shortcuts?
19:05
ooh yeah that rings a bell, so it would be in the subclassing code
19:35
I searched but I only see EasyHook referenced in either Rubberduck.Main and in Rubberduck.Core so I guess the subclassing code doesn't depend on the EasyHook.
 
3 hours later…
23:53
[rubberduck-vba/Rubberduck] 1699 stars vs. [decalage2/oletools] 2301 stars

« first day (3138 days earlier)      last day (42 days later) »