VBA Rubberducking

Duga

12:00 AM

RELOAD!

[rubberduck-vba/Rubberduck] 4 issue comments.

Peter M Taylor

1:27 AM

@Vogel612 as in YouTube channels or builds thru appvoyer verison checking?

Vogel612

neither

Peter M Taylor

Oh

Peter M Taylor

2:01 AM

Hope this does not complicates GH policy blog.github.com/2018-03-14-eu-proposal-upload-filters-code

Vogel612

2:20 AM

would be a bit of a pain ...

either way... 3:30 AM. I should get to bed

ThunderFrame

2:46 AM

@FreeMan except each workbook can have it's own ribbon defined, or have a different tab active.

@FreeMan huh, "Looping Mucho Slowo" seems to translate to Chinese easily too: "Loo Ping Mu Cho Slo Wo"

this

3:01 AM

@MathieuGuindon does this make sense to you?

+; TODO this CLSID should match the one used by the current installer.
+AppId={{979AFF96-DD9E-4FC2-802D-9E0C36A60D09}

ThunderFrame

cowers in over-use of apostrophe shame

this

that's 3 years ago, and nobody has commented on it, I think.

@ThunderFrame but is it valid Chinese?

No hablo chino.

ThunderFrame

68% ~~of statistics are made up~~ likely

this

LOL

suppose that's about true of about any languages.

ThunderFrame

@this except LISP

this

3:07 AM

Well, LISP is easy. Just use enough parentheses.

and make sure you lisp as you parenthesize.

this

5:02 AM

2

Duga

3:19 PM

[rubberduck-vba/Rubberduck] bclothier synchronized pull request #3836: Gather metadata required for per-user install for Rubberduck

[rubberduck-vba/Rubberduck] bclothier commented on pull request #3836: Gather metadata required for per-user install for Rubberduck

> The latest commits now will build a new installer, with all the parameterization taken in. However, the installer isn't fully functional and running it will cause the VBA to not load. More troubleshooting is needed.

Note that the installer script has been revised a... lot in order to allow for better parameterization and minimizing the code required. I have not cleaned up the code that should be no longer used (e.g. the one for performing the `regasm.exe` - those are no longer called any m

[rubberduck-vba/Rubberduck] build for commit 2e841e1f on unknown branch: AppVeyor build failed

BUILD FAILURE!

[rubberduck-vba/Rubberduck] build for commit 2e841e1f on unknown branch: 55.99% (target 0%)

Duga

4:25 PM

[rubberduck-vba/Rubberduck] bclothier synchronized pull request #3836: Gather metadata required for per-user install for Rubberduck

Mathieu Guindon

@this OMG!!!

We can't merge a non-working / WIP installer script though

But this is very exciting!

Hosch250

So, if I understand correctly, we don't need develop RD with admin access for the registry anymore?

Because Wayne got us the stuff for running unit tests directly?

@MathieuGuindon Your sister doing any better?

Duga

[rubberduck-vba/Rubberduck] build for commit 0880a60d on unknown branch: AppVeyor build failed

BUILD FAILURE!

[rubberduck-vba/Rubberduck] build for commit 0880a60d on unknown branch: 55.99% (target 0%)

this

@Hosch250 no, that's a separate thing

it always was possible to do per-user installation.

it's just that the tools like regasm.exe is too dumb

Hmm.

Exception calling "WriteAllLines" with "3" argument(s): "Could not find a part of the path 'C:\projects\rubberduck\Rubberduck.Deployment\InnoSetup\Includes\Rubberduck.reg.‌iss'."

@MathieuGuindon Of course not. I fully anticipate that this will have to be tested by lot of other people besides me to know it in fact works on a variety of Windows, blah blah blah blah

Mathieu Guindon

4:44 PM

@Hosch250 yeah. her immune system is slowly recovering. we're currently playing scrabble over FB messenger lol

Hosch250

@MathieuGuindon Awesome.

So, it's incurable, but is it possible that she'll be able to live a mostly-normal life if she recovers enough?

Duga

[rubberduck-vba/Rubberduck] bclothier synchronized pull request #3836: Gather metadata required for per-user install for Rubberduck

this

^ I think it's all because of git not including empty folders. Let's see if it does the trick.

Mathieu Guindon

@Hosch250 yeah. she's even (slowly) going back to work at the end of the month

Hosch250

That's awesome.

Duga

5:02 PM

[rubberduck-vba/Rubberduck] build for commit d0c824b0 on unknown branch: AppVeyor build failed

BUILD FAILURE!

[rubberduck-vba/Rubberduck] build for commit d0c824b0 on unknown branch: 55.99% (target 0%)

this

Very good to hear, @MathieuGuindon.

2

@Duga such is the joy of building on a remote system.

This one has me stumped; it builds locally and the IS version are the same.

Vogel612

5:22 PM

@this the installer script fails

the build is fine, the installer script is a bit annoyed about AppVersion / AppVerName

> Error in C:\projects\rubberduck\Rubberduck.Deployment\InnoSetup\Installer Build Script.iss: The [Setup] section must include an AppVersion or AppVerName directive.

Duga

5:35 PM

[rubberduck-vba/Rubberduck] IvenBach commented on issue #3843: Unit Test inconclusive with long path

> Rubberduck version: Version 2.1.6642.26611
Operating System: Microsoft Windows NT 6.1.7601 Service Pack 1, x64
Host Product: Microsoft Office 2013 x86
Host Version: 15.0.5007.1000
Host Executable: EXCEL.EXE

this

6:45 PM

@Vogel612 yes but i was referring to the IS bulidng locally as well. I think theres a problem with the expansion of the #define constant up there but Im noy sure why its ok here but not there.

Vogel612

8:38 PM

@this I'll see whether I can take a look tomorrow

Hosch250

8:54 PM

@FreeMan RD settings have a significantly more complicated UI than that.

Duga

9:34 PM

[rubberduck-vba/Rubberduck] rkapka created pull request #3845: Fixed inspections and quick fixes registration in Castle Windsor to merge rkapka/Rubberduck/rkapka-ioc into rubberduck-vba/Rubberduck/next

> Apparently some PR broke the IoC registration for inspections and quick fixes. Only ParseTreeInspections were registered correctly. Other inspections and all quick fixes were not registered at all or registered having service type of Object. This PR's purpose is to restore the functionality.

[rubberduck-vba/Rubberduck] rkapka commented on pull request #3845: Fixed inspections and quick fixes registration in Castle Windsor

> I'm not familiar with CW, so this fix might not be the best approach. I don't know why Base and Self services were not working.

[rubberduck-vba/Rubberduck] rkapka review_requested pull request #3845: Fixed inspections and quick fixes registration in Castle Windsor

[rubberduck-vba/Rubberduck] build for commit b5a7c477 on unknown branch: AppVeyor build succeeded

[rubberduck-vba/Rubberduck] [codecov[bot]](github.com/apps/codecov) commented on pull request #3845: Fixed inspections and quick fixes registration in Castle Windsor

> # [Codecov](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/3845?src=pr&el=h1) Report
> Merging [#3845](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/3845?src=pr&el=desc) into [next](https://codecov.io/gh/rubberduck-vba/Rubberduck/commit/443a039b294800b97b6061747ebb98b61cc1a160?src=pr&el=desc) will **increase** coverage by `0.01%`.
> The diff coverage is `100%`.

```diff
@@ Coverage Diff @@
## next #3845 +/- ##
===========================

[rubberduck-vba/Rubberduck] build for commit b5a7c477 on unknown branch: 56% (target 0%)

Hosch250

9:50 PM

@this Remember our discussion about defaults a while back?

I figured out the solution.

We need to populate the defaults in the new() ctor.

Because the ctor runs before the values are dumped in.

Duga

[rubberduck-vba/Rubberduck] retailcoder merged pull request #3845: Fixed inspections and quick fixes registration in Castle Windsor from rkapka/Rubberduck/rkapka-ioc into rubberduck-vba/Rubberduck/next

[rubberduck-vba/Rubberduck] rkapka pushed commit b5a7c477 to next: Fixed inspections and quick fixes registration in Castle Windsor

[rubberduck-vba/Rubberduck] web-flow pushed commit 7b74b4b4 to next

Merge pull request #3845 from rkapka/rkapka-ioc

Emergency fix restoring inspections and quickfix functionalities.

Hosch250

@Vogel612 Just how much did you break? ;)

Reminds me of some of my escapades at work.

Actually, if we initialize the default values there, we don't even need the "default" settings file--just new them up.

Duga

[rubberduck-vba/Rubberduck] retailcoder published prerelease Rubberduck-v2.1.2.2964

[rubberduck-vba/Rubberduck] retailcoder created tag Rubberduck-v2.1.2.2964

[rubberduck-vba/Rubberduck] build for commit 7b74b4b4 on next: AppVeyor build succeeded

[rubberduck-vba/Rubberduck] build for commit 7b74b4b4 on next: 56% (target 0%)

Vogel612

10:16 PM

@Hosch250 that directly means we have the defaults in code again.

Then again that's irrelevant. We're only changing them with a binary change anyways, soo ...

Hosch250

Correct. But does it matter? Because this way we can see everywhere we populate them.

And that ^

And we need to distribute the "defaults" somehow anyway...

Currently we are just shipping the file with RD, and I see no other way given the nature of the project, really. It's not like we are pulling the defaults from somewhere online each time they open RD.

Vogel612

@Hosch250 I didn't actually touch these lines themselves... maybe Where and BasedOn behave differently for registration purposes

that said... UnitTest there might have helped catch that problem ...

Duga

10:52 PM

[rubberduck-vba/Rubberduck] MDoerner commented on Rubberduck.Main/Root/RubberduckIoCInstaller.cs of pull request #3845: Fixed inspections and quick fixes registration in Castle Windsor

[rubberduck-vba/Rubberduck] MDoerner commented on pull request #3845: Fixed inspections and quick fixes registration in Castle Windsor

> I just looked further in the CW documentation. There was no link in the part about registration by convention but in the part about registering component-by-component.

[Conditional component registration](https://github.com/castleproject/Windsor/blob/master/docs/conditional-component-registration.md)

After reading this, I think we should go back to `BasedOn` combined with `If` or `Unless` to deal with the experimental types.

Hosch250

I don't think anyone can compare to Ansel Adams even today, but I think a significant part of the power of his work is the medium.

Most of his pictures wouldn't be portrayed so powerfully if they didn't take advantage of the starkness of black & white medium he used.

Some would be more beautiful, but sometimes I think cutting the color away helps show us the underlying object a little better.

Of course, you can't just look at the b&w image and say "now I know {x}". You need to understand it in color too.

this

@Hosch250 that's basically what my PR did.

Hosch250

@this Correct, but wrong ctor.

Oh wait, you did it on the setting ctor?

I was thinking it was the view model ctor for the settings window.

Yeah, good job.

this

11:08 PM

yeah, got me second guessing for a second. :)

Hosch250

Yeah, I was way too tired to be reviewing that night.

Duga

[rubberduck-vba/Rubberduck] Hosch250 closed issue #3795: New setting CompileBeforeParse is not defaulting to true

this

No worries. I appreciate the 2nd set of eyes

Duga

[rubberduck-vba/Rubberduck] Hosch250 merged pull request #3840: Enforce the non-default default values from bclothier/Rubberduck/FixSettingDefault into rubberduck-vba/Rubberduck/next

[rubberduck-vba/Rubberduck] bclothier pushed commit 4810e3fc to next: Enforce the non-default default values

[rubberduck-vba/Rubberduck] bclothier pushed commit 352e72ac to next: Merge branch 'next' of github.com/rubberduck-vba/Rubberduck into FixSettingDefault

[rubberduck-vba/Rubberduck] web-flow pushed commit 7983a707 to next

Merge pull request #3840 from bclothier/FixSettingDefault

Enforce the non-default default values

[rubberduck-vba/Rubberduck] Hosch250 commented on pull request #3825: Fix for the Remove Parameter Refactoring

> @Vogel612 github.com/MDoerner/Rubberduck/blob/…

[rubberduck-vba/Rubberduck] Hosch250 merged pull request #3825: Fix for the Remove Parameter Refactoring from MDoerner/Rubberduck/FixForRemoveParameter into rubberduck-vba/Rubberduck/next

[rubberduck-vba/Rubberduck] Hosch250 pushed 12 commits to next (only showing some of them below)

[rubberduck-vba/Rubberduck] MDoerner pushed commit 29bb1612 to next: Removed useless COM accesses from remove parameter refactoring.

[rubberduck-vba/Rubberduck] MDoerner pushed commit f50d522c to next: Added failing tests for removing optional parameters supplied as missing parameters on call sites.

[rubberduck-vba/Rubberduck] MDoerner pushed commit 004efb4d to next: Added failing test for remove parameter removing the argument from call sites in other components.

[rubberduck-vba/Rubberduck] MDoerner pushed commit 1e10cd0c to next: Fixed rewriting callers in other components in the remove parameter refactoring.

[rubberduck-vba/Rubberduck] MDoerner pushed commit b20b017c to next: Removed projects provider from remove parameter refactoring since it is no longer needed.

[rubberduck-vba/Rubberduck] MDoerner pushed commit 6c90dc4f to next: Corrections to expected values in failing remove parameter tests for missing arguments.

[rubberduck-vba/Rubberduck] MDoerner pushed commit e8c2937c to next: Fixed removal of missing arguments for remove parameter refactoring.

[rubberduck-vba/Rubberduck] MDoerner pushed commit 58a5da93 to next: Merge branch 'next' into FixForRemoveParameter

[rubberduck-vba/Rubberduck] MDoerner pushed commit 57ebbca2 to next: Merge branch 'next' into FixForRemoveParameter

[rubberduck-vba/Rubberduck] web-flow pushed commit e2def24f to next

Merge pull request #3825 from MDoerner/FixForRemoveParameter

Fix for the Remove Parameter Refactoring

[rubberduck-vba/Rubberduck] bclothier synchronized pull request #3836: Gather metadata required for per-user install for Rubberduck

[rubberduck-vba/Rubberduck] retailcoder published prerelease Rubberduck-v2.1.2.2965

[rubberduck-vba/Rubberduck] retailcoder created tag Rubberduck-v2.1.2.2965

[rubberduck-vba/Rubberduck] build for commit 7983a707 on next: AppVeyor build succeeded

[rubberduck-vba/Rubberduck] build for commit 7983a707 on next: 56.01% (target 0%)

this

@Duga this is expected to fail building. I enabled verbose output to help diagnose why AV won't work the way my local works.

Hosch250

11:20 PM

@this Perhaps missing dependencies?

Or perhaps AppVeyor is blocking registry reads?

this

unlikely

I did get the same error myself when I was working locally

this is because if it's not valid, the version get left empty, which isn't legal

this is all on this line: #define BuildDir ExtractFileDir(ExtractFileDir(SourcePath)) + "\bin\Debug\"

conceptually equivalent to doing a `..\..\bin\Debug`

that's why the PR - there are #pragma message that will output the result of the expression.

Duga

[rubberduck-vba/Rubberduck] retailcoder published prerelease Rubberduck-v2.1.2.2966

[rubberduck-vba/Rubberduck] retailcoder created tag Rubberduck-v2.1.2.2966

[rubberduck-vba/Rubberduck] build for commit e2def24f on next: AppVeyor build succeeded

[rubberduck-vba/Rubberduck] build for commit e2def24f on next: 56.02% (target 0%)

[rubberduck-vba/Rubberduck] build for commit be107c31 on unknown branch: AppVeyor build failed

BUILD FAILURE!

[rubberduck-vba/Rubberduck] [codecov[bot]](github.com/apps/codecov) commented on pull request #3836: Gather metadata required for per-user install for Rubberduck

> # [Codecov](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/3836?src=pr&el=h1) Report
> Merging [#3836](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/3836?src=pr&el=desc) into [next](https://codecov.io/gh/rubberduck-vba/Rubberduck/commit/443a039b294800b97b6061747ebb98b61cc1a160?src=pr&el=desc) will **increase** coverage by `0.03%`.
> The diff coverage is `89.47%`.

```diff
@@ Coverage Diff @@
## next #3836 +/- ##
=======================

[rubberduck-vba/Rubberduck] build for commit be107c31 on unknown branch: 56.02% (target 0%)

this

11:38 PM

so the path is ok. I was looking at the wrong #define

the real culprit is #define AppVersion GetFileVersion(BuildDir + "Rubberduck.Core.dll")

Tweeting Duck

Moderator elections coming up on Code Review Stack Exchange. Thank you for your many years of service, 200_success! https://twitter.com/StackCodeReview/status/975148821931012098

Tweeted by rubberduckvba on March 17, 2018 at 11:32 PM

Not sure if this is sad, or genius... https://twitter.com/kylebshr/status/973965922670018560

Tweeted by rubberduckvba on March 17, 2018 at 11:38 PM

Duga

[rubberduck-vba/Rubberduck] retailcoder added label PR-Status: Conflicting to pull request #3835: Freeze column header for vertical scrolling.

[rubberduck-vba/Rubberduck] retailcoder merged pull request #3823: Unify some duplicated codes and make them testable/mockable from bclothier/Rubberduck/UnifyVbeFunctions into rubberduck-vba/Rubberduck/next

[rubberduck-vba/Rubberduck] retailcoder pushed 24 commits to next (only showing some of them below)

[rubberduck-vba/Rubberduck] bclothier pushed commit 489b073c to next: Fix the check on the UiDispatcher; it does not need the CheckContext.

[rubberduck-vba/Rubberduck] bclothier pushed commit 9ef93be3 to next: Make ComMessagePumper take IVBERuntime as an injected dependency instead of newing it up.

[rubberduck-vba/Rubberduck] bclothier pushed commit af33e828 to next: Correct the logic for checking the thread in the pumper.

[rubberduck-vba/Rubberduck] bclothier pushed commit fcd3dca0 to next: Remove unused using declaration

[rubberduck-vba/Rubberduck] bclothier pushed commit 8575191b to next

Merge branch 'next' of https://github.com/rubberduck-vba/Rubberduck into UnifyVbeFunctions

# Conflicts:
# RetailCoder.VBE/VBERuntime/IVBESettings.cs
# RetailCoder.VBE/VBERuntime/RegistryWrapper.cs
# RetailCoder.VBE/VBERuntime/VBESettings.cs
# Rubberduck.Core/Rubberduck.Core.csproj
# Rubberduck.Core/UI/Settings/SettingsForm.cs
# Rubberduck.Core/VBERuntime/IVBESettings.cs
# Rubberduck.Core/VBERuntime/RegistryWrapper.cs
# Rubberduck.Core/VBERuntime/VBESettings.cs
# Rubberduck.Main/Root/Rubberduc

[rubberduck-vba/Rubberduck] bclothier pushed commit b6cc24ed to next: Clean up the botched merge

[rubberduck-vba/Rubberduck] bclothier pushed commit 606f8288 to next: Address feedbacks and restore some missing code from merge.

[rubberduck-vba/Rubberduck] bclothier pushed commit 195684d7 to next: Make VBEEvents mockable, thanks to Mug

[rubberduck-vba/Rubberduck] bclothier pushed commit 1a5e59f1 to next: Make the guard clauses in RPS explicit, add ignore attribute. Add explanatory comment to highlight the static methods being used for the VBEEvents.

[rubberduck-vba/Rubberduck] web-flow pushed commit 47af462d to next

Merge pull request #3823 from bclothier/UnifyVbeFunctions

Enhances testability.

[rubberduck-vba/Rubberduck] retailcoder published prerelease Rubberduck-v2.1.2.2968

[rubberduck-vba/Rubberduck] retailcoder created tag Rubberduck-v2.1.2.2968

[rubberduck-vba/Rubberduck] build for commit 47af462d on next: AppVeyor build succeeded

[rubberduck-vba/Rubberduck] build for commit 47af462d on next: 55.98% (target 0%)

Transcript for