VBA Rubberducking

Mathieu Guindon

11:00 PM

was SC, but the export feature was invoked by SC

so, there's a chance it's still there

this

I'm currently testing this

2

Mathieu Guindon

#RecursiveTesting

Comintern

Would be more appropriate on StackOverflow.

Duga

[rubberduck-vba/Rubberduck] retailcoder added label stale to issue #2315: Init Repo doesn't carry User Form Object Blob

[rubberduck-vba/Rubberduck] retailcoder closed issue #2315: Init Repo doesn't carry User Form Object Blob

[rubberduck-vba/Rubberduck] retailcoder is playing around with a project card

[rubberduck-vba/Rubberduck] bclothier created comment on issue #2299: Parens () are added into the code when comitting

> Cannot reproduce, closing. Exporting either a standard module or a document module does not produce this.

[rubberduck-vba/Rubberduck] bclothier closed issue #2299: Parens () are added into the code when comitting

[rubberduck-vba/Rubberduck] bclothier is playing around with a project card

Mathieu Guindon

@Duga @this #2343 seems related

Duga

11:02 PM

[rubberduck-vba/Rubberduck] bclothier added label status-norepro to issue #2299: Parens () are added into the code when comitting

[rubberduck-vba/Rubberduck] bclothier added label stale to issue #2299: Parens () are added into the code when comitting

this

already on it

Duga

[rubberduck-vba/Rubberduck] retailcoder added label stale to issue #2472: Interacting with ToolWindows disables their dockability for the remainder of the VBE session

[rubberduck-vba/Rubberduck] retailcoder created comment on issue #2472: Interacting with ToolWindows disables their dockability for the remainder of the VBE session

> Presumably fixed.

Mathieu Guindon

@Duga right?

Duga

[rubberduck-vba/Rubberduck] retailcoder closed issue #2472: Interacting with ToolWindows disables their dockability for the remainder of the VBE session

[rubberduck-vba/Rubberduck] retailcoder is playing around with a project card

this

Don't know. I tried to add the button then I.... lost it.

Mathieu Guindon

11:05 PM

?

this

in the 2nd gif, you can see Thunderframe customizing the toolbars by adding the Dockable button.

I dragged it to my toolbar and.... poof

can't get it back

hm

it's gone.

so, not fixed then

that's not RD

that's VBE's button

so I can't test it properly

Mathieu Guindon

11:08 PM

ttqw

Duga

[rubberduck-vba/Rubberduck] rkapka created comment on issue #4707: Rubberduck programming language

> > Exactly what would get removed? To do what?

All the VBA constructs that are obsolete, confusing or not useful. Maybe #3486 or #3308 could be such examples? I see a benefit of not having to write inspections for all such constructs because they will just not compile. RD would document what constructs were removed and why. And if the programmer *really* needs to use a removed construct, they can switch back to VBA.

> I originally read the suggestion as to write a transpiler

Yes, I was

this

hmm RD's import doesn't overwrite.

i think it'd be nice if it supported overwriting, instead of adding a xxx1

KySoto

@this oh its a network copy of a database. i built a system to version access apps and download a copy to their computer

it totally solves the issue.... except when the users dont use the powershell script shortcut i put on their computers

Comintern

Alright, time to merge next.

this

ah oof.

KySoto

11:15 PM

yeah

this

we had a similar situation - the fix was to add a startup code to check the path. If it's a network path, quit immediately

KySoto

i built this app like almost 3 years ago

that i can use to track all of our access applications

and what database tables they use

it ended up morphing into our application tracking

so we do changelogs for apps in it

and with that came the whole powershell script that looks up to see what the last revision you ran was and compares it to the most recent changelog entry

Duga

[rubberduck-vba/Rubberduck] bclothier created comment on issue #4707: Rubberduck programming language

> Ok, I think I see what you mean -- it sounds more like you want something equivalent to JavaScript's `'use strict';`? Force all obsolete constructs to be compile-time errors?

Consider that one goal of Rubberduck is to enable developers to take a legacy hand-me-down codebase with all wrong stuff and crappy constructs and be able to maintain it... especially without breaking it. Enabling a strict mode would not help in the initial stage but as the cleanup progressing, I can see enabling stric

KySoto

then when its done, if it was running a different version then the most recent change, it downloads a new copy of the app to the computer, and then logs it in the table storing what versions of applications the computers have

it also logs when that update happened plus every time the script runs, it updates the last ran field, so i know if people are actually using the shortcut

its a neat solution for our setup

the best part is, i dont have to interupt users of the application when i need to update the application since it checks every time it runs

now that does put a damper on things when they leave it up for several months, and we have updated it a bunch

Comintern

OK, sanity check. Do I need to squash commits after rebasing my inspection branch, or will the PR merge do that?

this

11:22 PM

when I made a similar mistake, merging did the trick

KySoto

i can guarantee you that it will do things that i am not familiar with.

Duga

[rubberduck-vba/Rubberduck] bclothier created comment on issue #2343: empty lines added to all files on Source Control Refresh

> I cannot reproduce this so closing it.

[rubberduck-vba/Rubberduck] bclothier closed issue #2343: empty lines added to all files on Source Control Refresh

[rubberduck-vba/Rubberduck] bclothier did something with some project card

[rubberduck-vba/Rubberduck] bclothier added label status-norepro to issue #2343: empty lines added to all files on Source Control Refresh

[rubberduck-vba/Rubberduck] bclothier added label stale to issue #2343: empty lines added to all files on Source Control Refresh

this

hooray, we no longer have 7th page of bugs. :-p #ArbitaryMeasurement

2

Comintern

OK, that was the missing piece. The merge pulled in InvalidateStaleInspectionResults, and it needs a cast now.

Mathieu Guindon

@Comintern we can squash-and-merge, but we'd rather not do that

wow tonight's day-end summary will be interesting

Comintern

11:25 PM

Nah, I think we're good.

this

huh. that's new.

Mathieu Guindon

@this wth?

Duga

[rubberduck-vba/Rubberduck] retailcoder created comment on issue #4707: Rubberduck programming language

> I think that's an interesting idea... but well out of scope =)

this

IKR? I'm trying to find out who is giving me that. I was closing Excel and Access, Access crashed while closing and is already dead.

Mathieu Guindon

@Duga not closing to keep discussion going, but IMO that's well beyond RD's scope

like, waay beyond

this

11:31 PM

hmm, belongs to desktop

welp, i have no idea.

Comintern

Crap.

> Updates were rejected because the tip of your current branch is behind its remote counterpart.

I don't think I want to pull in this situation though. Do I need to git push --force?

this

I think that's what I did

you're talking about your fork as the origin, right?

Comintern

No, wait. I think I do want to pull.

this

pull from rd's next, yes. From your fork, never.

Comintern

Yes. From the standpoint of local/inspections

I'm already based even with RD's next.

Duga

11:36 PM

[rubberduck-vba/Rubberduck] bclothier closed issue #2493: Hotkeys only register in 1 application if multiple RD instances are running.

[rubberduck-vba/Rubberduck] bclothier did something with some project card

[rubberduck-vba/Rubberduck] bclothier added label status-norepro to issue #2493: Hotkeys only register in 1 application if multiple RD instances are running.

Comintern

I think I just need the history from the commits that were pushed before the rebase.

Duga

[rubberduck-vba/Rubberduck] bclothier added label stale to issue #2493: Hotkeys only register in 1 application if multiple RD instances are running.

Vogel612

@Comintern neither

@Comintern depends... want to completely ignore what's on remote and kill it with fire? that's what force-push is for

if that message appears, the branches are "divergent"

if you want changes from both branches, you'll need to merge or rebase

Comintern

I'm trying to remember how many commits there were and whether I care about preserving them.

Vogel612

git pull --rebase is a thing

git log --oneline --decorate --graph --all

Duga

11:42 PM

[rubberduck-vba/Rubberduck] Vogel612 added label status-declined to issue #4707: Rubberduck programming language

[rubberduck-vba/Rubberduck] Vogel612 closed issue #4707: Rubberduck programming language

Comintern

OK, this seems to be where it's messed up.

| * | | | | | | dc9039cb0 Increase performance of GetAnnotatedLine
| * | | | | | | 84cf953b3 Scope annotations to next non-whitespace line
| | | | | | | | * c9cd8a77e (origin/inspections) Stop churning results on filter and group, freshen up toolbar.
| | | | | | | |/
| | | | | | |/|
| | | | | | * | d84f3e76a Use static images and converters.
| | | | | | |/
| | | | | | * 5cf64236a Apply CE refactoring to CodeMetrics window.
| | | | | | * 4ad676e0f Fix test setup.
| | | | | | * 0a22f2508 Code Explorer performance rewrite.

Vogel612

looks like a candidate for pull --rebase if HEAD points to the far right line

Comintern

Yeah, I think pull should be fine.

Vogel612

just pulling will make git go a bit cross-eyed, tho

because possible merge conflicts

Duga

[rubberduck-vba/Rubberduck] bclothier closed issue #2804: Code Inspections are left over when a workbook is closed.

[rubberduck-vba/Rubberduck] bclothier created comment on issue #2804: Code Inspections are left over when a workbook is closed.

> Appears to be been fixed already. Closing.

[rubberduck-vba/Rubberduck] bclothier moved project card