VBA Rubberducking - 2018-10-17

« first day (1594 days earlier) ← previous day next day → last day (1586 days later) »

12:01 AM

RELOAD!

[bruglesco/fleet-command] 8 commits. 1 opened issue. 1 closed issue. 722 additions. 87 deletions.

[rubberduck-vba/Rubberduck] 1 opened issue. 9 issue comments.

[Zomis/zomis.github.io] 2 commits. 2448 additions.

[Zomis/zomis.github.io-source] 9 commits. 102 additions. 25 deletions.

12:19 AM

[rubberduck-vba/Rubberduck] retailcoder commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

[rubberduck-vba/Rubberduck] retailcoder commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

Mathieu Guindon

@IvenBach you're doing great. I'd try to reuse as much of the existing annotation-related code though, but the general approach is fine IMO.

Trying to pay back the knowledge ya'll helped me finally connect the dots on.

5 hours later…

5:43 AM

@this I could help out in Powershell whenever I could. At least gain back your sanity in some form.

But I don’t have azure connection as I finished my trial didn’t feel the need to continue

@Vogel612 I see the network dependency of 36 modules on mine iPad. Like this said, shows how other people depend on us and which version related. Interesting shows how deep .net is accessed. Slightly misleading saying vulnerability however I imagine down the track .net may issue alerts on their security affecting us by how much who knows at this time.

3 hours later…

8:31 AM

[rubberduck-vba/Rubberduck] MDoerner commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

[rubberduck-vba/Rubberduck] MDoerner commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

[rubberduck-vba/Rubberduck] MDoerner commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

[rubberduck-vba/Rubberduck] MDoerner commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

[rubberduck-vba/Rubberduck] MDoerner commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

[rubberduck-vba/Rubberduck] MDoerner commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

[rubberduck-vba/Rubberduck] MDoerner commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

[rubberduck-vba/Rubberduck] MDoerner commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

[rubberduck-vba/Rubberduck] MDoerner commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

[rubberduck-vba/Rubberduck] MDoerner commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

[rubberduck-vba/Rubberduck] Vogel612 commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

[rubberduck-vba/Rubberduck] snoozins starred us

8:49 AM

Alright folks: if within 36 hours of now I don't get any reviews or objections, I will merge the csproj update PR

It's been ready for review since three days already

2 hours later…

10:57 AM

I will definitely not review since I have no idea about the project format, neither the new nor the old.

1 hour later…

12:23 PM

[rubberduck-vba/Rubberduck] comintern commented on Rubberduck.Core/UI/CodeExplorer/Commands/ImportCommand.cs of pull request #4444: Issue4282 import module folder annotation

12:38 PM

@Vogel612 just confirming - the issue about UI being broken was addressed, right?

12:56 PM

Yes.

It was symptomatically fixed by affixing the version in Rubberduck.Core to something without a wildcard

Right. I will have a quick review tonight.

1:13 PM

@Vogel612 Any particular reason that inspection/quick fix was removed?

EmptyConditionBlockInspection

Other than that, it looks good.

That one was replaced by more specific quickfixes and while the file was removed from the csproj specifications it was not deleted

It says deleted. github.com/rubberduck-vba/Rubberduck/pull/4383/…

Compare specifically: github.com/rubberduck-vba/Rubberduck/blob/next/…

There's a handful of others as well

And I don't see any new inspection files; just this deleted one.

Yea that's happened quite a while ago

Check the csproj file from current next whether you can find the deleted file referenced in there

1:25 PM

OK. Looks fine then, if this is what you are expecting.

Someone split that inspection ages ago and missed that file

Oh, makes sense now.

So, that file was removed from the solution, but not the source control.

They removed it from the project, but not from version control

We have some of that at work too. More than I like.

Typing on mobile is so damn slow

The globs just make stuff like that show up

1:27 PM

You have a couple merge conflicts.

Gee... again?

Again.

Who merged stuff again? I can take care of those when I get home later tonight.

I'm not sure; not me.

At least that kind of merge conflict will disappear when the PR is merged

And all the other PRs that are currently open also get their merge-conflicts resolved 😅

1:30 PM

:)

I'm hoping to get work moved over soon.

But not being an "architect" will make it very difficult.

But the good news is it looks like I'm on track to get recognized as a top performer in the company, and that comes with a nice bonus of points in the company store.

At least some files was from my PR that Mat merged. Should not be a big deal.

Nice, Hosch!

1:43 PM

Probably, there are also some file changes from one of my last PRs.

Did I get that right, once the PR is merged, renaming/moving/deleting/adding files will no longer cause merge conflicts on the project files?

@M.Doerner Correct.

Although, all PRs made with the old version will probably have massive conflicts.

additionally, we now have a RubberduckBaseProject to provide a standard setting for all other projects.

2:04 PM

morning

2:28 PM

> But then, have you ever cut your hand on a sheet of paper?

What's paper? :P

Actually, that should've been "wuts papr".

@M.Doerner thank you for the descriptive comments. I’ll have questions about them when I start updating the PR later today.

2:44 PM

@Vogel612 are the targets guaranteed to be executed in same sequential order as listed in the csproj?

Someone on SE just offended me. They are recommending polluting the ocean!

> Pack up all your computers and throw them in the ocean.

@this no. The targets are guaranteed to follow the order they require though. BeforeTargets and AfterTargets attributes are the factor here

Everything else (PropertyGroup and Items [Compile/Resource/...]) is guaranteed to be run in encounter order

That's especially true for the import of the base project

ok, and is there any reason why the copyright isn't centralized in the RubberduckBaseProject?

@Hosch250 what douche kabobs

3:34 PM

@Hosch250 It's bound to happen. Plenty of divergent idears on teh interweebz.

It was mostly a joke :P

I wasn't really offended, it was about someone concerned about security barking up the wrong tree, and refusing to realize it.

And the person they were with was like "fine, if you think this is a concern, yes, it's possible, but so unlikely that you might as well never use a computer again if you think it's something to get concerned over because all the other issues are bigger."

3:56 PM

[rubberduck-vba/Rubberduck] SanAndTang opened issue #4445: Cannot get to Settings in version 2.2.0

> On the Rubberduck menu, the Settings submenu does not open the settings window.

Dupe, IIRC.

[rubberduck-vba/Rubberduck] k-j-kleist created comment on issue #4445: Cannot get to Settings in version 2.2.0

> The same here, since I updated one (or two?) days ago:
```
Version 2.2.0.4008
OS: Microsoft Windows NT 10.0.17134.0, x64
Host Product: Microsoft Office 2016 x86
Host Version: 16.0.9126.2295
Host Executable: EXCEL.EXE
```

4:11 PM

@this it isn't the same for all projects

Though a follow up could generate the current year part

Yeah I saw, but does it have to not be? Seems easier to just have it in one place.

The problem is where do we start asessing copyright

Well, Rubberduck.Deployment automate this for the license rtf file. So we're not that far away.

just use 2014-20XX

Hmmm...

if it's in only one place, then it becomes easier to manage the year, too.

4:17 PM

#TIL F10 doesn't do what I want it to in VBA for code execution...

My keyboard goes to F11.

@Comintern So sadz 4 u. I haz a F12.

:waits to see if anyone else gets the reference:

4:38 PM

[rubberduck-vba/Rubberduck] k-j-kleist edited comment on issue #4445: Cannot get to Settings in version 2.2.0

> The same here, since I updated one (or two?) days ago:
```
Version 2.2.0.4008
OS: Microsoft Windows NT 10.0.17134.0, x64
Host Product: Microsoft Office 2016 x86
Host Version: 16.0.9126.2295
Host Executable: EXCEL.EXE
```
And by "updated", I mean:
- Exit Excel
- Check in Task Manager that is was no longer running
- From Window Settings > App, uninstall Rubberduck (2.2.0.3971)
- Install the new version (in the same folder where the old version had lived)

[rubberduck-vba/Rubberduck] MDoerner created comment on issue #4445: Cannot get to Settings in version 2.2.0

> Thanks for reporting this.

I can reproduce the issue and I am currently investigating which changes caused it.

[rubberduck-vba/Rubberduck] MDoerner added label bug to issue #4445: Cannot get to Settings in version 2.2.0

[rubberduck-vba/Rubberduck] MDoerner added label critical to issue #4445: Cannot get to Settings in version 2.2.0

[rubberduck-vba/Rubberduck] MDoerner added label regression to issue #4445: Cannot get to Settings in version 2.2.0

4:53 PM

[rubberduck-vba/Rubberduck] MDoerner created comment on issue #4445: Cannot get to Settings in version 2.2.0

> This issue has been introduced with PR #4217.

5:16 PM

@Duga @M.Doerner When you learn why this happened can you let me know? I feel bad it was on my PR.

It's happened to us all. I've taken Prod down before.

Both in part and in whole.

You deleted it before.

No, that was just the next branch.

Prod on a website is much worse--it's the part clients actually work with.

That would be pretty bad.

I want to educate myself as to what happened to ~~avoid~~ minimize future regression issues.

besides it wasn't a green release

this is me shooting in dark but I think the settings is still too fragile

see, we had that same thing happen before (or a similar thing at least)

it does not take kindly to new elements being added to the setting file

or missing elements, so it goes PLUNK!

5:22 PM

I think it needs more brains on it.

I'll take another crack at it.

You know, Queen is really annoying, but Bohemian Rhapsody on sax sextet is pretty nice.

@this TBH, that's why I'm a bit stumped - that PR doesn't touch the settings files, only the UI.

@Comintern The UI probably got borked and is throwing an exception.

5:39 PM

@Hosch250 OK, found the source of my confusion. I could have sworn I'd been in settings since that PR was merged, but I synced just before it went in.

6:14 PM

[rubberduck-vba/Rubberduck] life2539 starred us

5

1 hour later…

7:23 PM

@PeterMTaylor ok - here's an example where I really get annoyed with posh. According to this documentation, the output is System.Management.Automation.PSRemotingJob. I set up a function to return this object and get an error saying that the type could not be found.

So I had to go to SO and find this thread which indicates that the type is an internal type and I should be using the base type System.Management.Automation.Job. How the hell am I supposed to know that from the documentation?

The problem is that this isn't just one thing - there's been several incidents like that so I end up dying by 10000 tiny cuts.

Didn't you feel like that when you started writing VBA too?

@Hosch250 IDK. I'm inclined to say no, because the help documentation were reasonably well written and I could reason my ways round.

With C#, the primary challenges were in new concepts that wasn't present in VBA

^

@this did you get started with VBA out of interest or needed to because of work?

With Posh, I already understand the concepts but those little things manage to bite me.

A little of both?

we used to do this stupid reporting in Excel. Used to be that we would take 1-2 days, closing the office to review the client casework to organize the required statistics.

At one point, I decided that I sucked at counting and gathering statistics, and built me a database.

Your skills stem from apathy then.

7:30 PM

That first Access db was fugly but it did work and the 1-2 days on a monthly basis went down to a 1 hour of running of report and filling out the excel spreadsheet by hand.

More like hatred, I'd say. I hated counting and tallying. I cannot stand mindnumbing menial task like that.

I'm writing an article for a co-worker that's having to do similar comparison stuff right now.

@IvenBach Laziness is a great motivational tool for programmers.

Console.WriteLine(GoodProgrammer.Equals(Lazy+HardWorking));

I've learned nearly everything I know about Excel because I didn't want to do it by hand.

8:22 PM

[rubberduck-vba/Rubberduck] Vogel612 synchronized pull request #4383: Csproj format update, fixes #3944

[rubberduck-vba/Rubberduck] codecov[bot] edited comment on pull request #4383: Csproj format update, fixes #3944

> # [Codecov](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/4383?src=pr&el=h1) Report
> Merging [#4383](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/4383?src=pr&el=desc) into [next](https://codecov.io/gh/rubberduck-vba/Rubberduck/commit/de6f99412db1492eb938dd94fa9ce43044716e5a?src=pr&el=desc) will **decrease** coverage by `6.9%`.
> The diff coverage is `n/a`.

```diff
@@ Coverage Diff @@
## next #4383 +/- ##
=============================

[rubberduck-vba/Rubberduck] build for commit dc7b5eb3 on unknown branch: AppVeyor build failed

BUILD FAILURE!

damn

possibly not my fault

8:44 PM

[rubberduck-vba/Rubberduck] build for commit dc7b5eb3 on unknown branch: AppVeyor build succeeded

yeap, not my fault. phew

9:10 PM

@this yeah I hear you mate. Another language with its own quirks. Although you provided me an example. I found some other projects using start-job in context, sometimes how they are used gives us something. github.com/sqlcollaborative/dbatools/blob/…

@this ^ the link code near the bottom appear to check the type and act accordingly.

thanks

[rubberduck-vba/Rubberduck] MDoerner created comment on issue #4445: Cannot get to Settings in version 2.2.0

> The origin of the problem is between the two merge commits cd379a103241515b5a1b7eea0cbaba5267e84187 and 3b94644619d9b2496815de3c3b4ea8ff2efb04a1, both syncing with next, probably one of the two. I cannot fully pinpoint the origin because the commits from the first sync till before the second one do not load at all. At the point of the first sync next was broken and did not load at all, since the merge of PR #4272.

@Duga We should really all remember to always check whether RD loads before opening/syncing a PR.

Moreover, when resolving a merge with conflicts, it would be nice to have the information about conflicting files in the commit message.

9:30 PM

@M.Doerner isn't that the default behavior?

I think so, but the second of the commits just has Resolve merge conflict as commit message.

Yes, I was not sure if that was the default git behavior because I use tortoise; I don't edit the comments for the merges normally and if I do, I usually append to it, rather than replacing it.

I also get them in the message, but I use SmartGit.

Ok, sounds like it's really git's default, which is good.

FWIW, if my theory about settings is correct, I would look at this: github.com/rubberduck-vba/Rubberduck/commit/…

Oh joy!

9:41 PM

Come to think of it... why are we enumerating code inspections in the settings....

Is there a way on GitHub to diff a merge commit against the previous commit from the branch merged in instead of the previous commit of the brach into which gets merged?

Figuring out which information is missing between two exports into excel without a primary key puts the FU in FUN.

@M.Doerner looks like it's here: github.com/rubberduck-vba/Rubberduck/compare

github.com/rubberduck-vba/Rubberduck/compare/…

Not sure that helps, though - there's apparently 145 commits between the 2 commits.

Thx

@M.Doerner github allows to diff between arbitrary commits...

if that helps..

9:49 PM

I did not find the corresponding page.

Now, I have bookmarked it.

I don't think there's an actual link

except maybe via create pull request?

why, it is!

very confusing. Nobody would think to go to New Pull Request if they just wanted to do a compare.

You do not have that many commits in between if you just diff to the other merge parent.

In that, the settings still showed up.

And it is only 13 file changes.

10:16 PM

so uhh what the eff

0

Q: Sikulix triggers runtime error 91 of a VBA analog macro?

the macro interface and sikulix codeThe macro incurs the runtime error 91 "Object variable or With block variable not set" is written in a special VBA environment constructed by Carl Zeiss microscope company. I use sikulix (sikulix.com) to type in the mask file and log file path with other numeri...

vba jython sikuli-x

i dont see any VBA in the picture that they linked

That's in the big ocean of SO. Stay within the breakwater of CR. At least there they have code you can review.

The big ocean is full of surprises and nasties you don't want to deal with.

[rubberduck-vba/Rubberduck] MDoerner created comment on issue #4445: Cannot get to Settings in version 2.2.0

> The problem seems to be that this line uses a resource image that no longer exists.

@IvenBach You need something else than the funnel for your filter button.

I erroneously removed it?

You referenced it in XAML and merged after somebody else had removed it.

10:29 PM

Ah. mkay.

Hm, what is the best way to find out where in the history a file has been deleted?

Getting the history for a file that is still there is easy, where do I start for the deleted one?

The image is actually there. Apprently, after resolving merge conflicts on the projects file, the image was no longer included in the project.

I'll open a hotfix PR for this.

so if I understand correctly, if a XAML file is missing a resource, the dialog basically blows up.

and this was apparently not caught during the usual build?

is there a way to force the XAML files to be fully compiled so we can verify all dependencies are correctly resolved?

Yes, if an image resource is missing, the constructor for the control throws.

I think compiling it does not even suffice.

You actually have to load it.

there might be a way to get some warnings generated from that

and those warnings in turn can be turned into errors...

Hm, still not showing.

More stuff seems to be borked.

10:45 PM

@Vogel612 if that's what it takes, then it should be done. The solution is far too large to let something like that slip by under the radar.

true

Would F5 to bring up the setting window catch it?

Ah, I forgot to set the build action to Resource.

@IvenBach well, you didn't bring up the settings window, did you?

^

10:47 PM

Yes, that would catch it.

I'd like to think that I did while testing...

However, we cannot really check everything manually all the time.

Still because there are several XAML files and those references several more resources (not just images)...

Besides it might had worked locally

(e.g. the file isn't in the solution/source control but is in the filesystem => works locally but breaks remotely)

Settings are back.

@this Where are we enumerating code inspections in the settings? IIR we used to inject them before the inspection refactor, but between storing the defaults in Settings.settings and refactoring the inspections themselves, that shouldn't be necessary anymore.

10:50 PM

it's in the settings.settings

Ah. Each inspection object used to be responsible for providing its default severity, which is needed to bootstrap the first settings XML file.

Otherwise, the first time you started RD, everything would be a warning level unless the concrete inspection specifically overrode it.

Yes and if I remember, sometime during your hiatus, this got changed

Yes. They used to be overrides on the concrete implementations IIR.

so i think it's actually ok. I was confused as I hadn't looked at the settings for a good while

[rubberduck-vba/Rubberduck] MDoerner created pull request #4446: Add Funnel.png to resource project as Resource to merge MDoerner/Rubberduck/FixSettingsNotShowing into rubberduck-vba/Rubberduck/next

> The addition of this resource had been lost when resolving merge conflicts and caused the settings window to throw and not open.

This PR closes #4445.

[rubberduck-vba/Rubberduck] Vogel612 created comment on pull request #4446: Add Funnel.png to resource project as Resource

> should be automatically fixed when merging #4383 as well.

10:54 PM

And now, TTGTB.

night.

@this I'd guess this is what may have occurred. #WorksOnMyBox but I never added it as part of PR.

I wonder if we should default the log level to trace on pre-release versions.

[rubberduck-vba/Rubberduck] build for commit a401e52f on unknown branch: AppVeyor build succeeded

[rubberduck-vba/Rubberduck] codecov[bot] created comment on pull request #4446: Add Funnel.png to resource project as Resource

> # [Codecov](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/4446?src=pr&el=h1) Report
> Merging [#4446](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/4446?src=pr&el=desc) into [next](https://codecov.io/gh/rubberduck-vba/Rubberduck/commit/de6f99412db1492eb938dd94fa9ce43044716e5a?src=pr&el=desc) will **decrease** coverage by `<.01%`.
> The diff coverage is `n/a`.

```diff
@@ Coverage Diff @@
## next #4446 +/- ##
==========================

[rubberduck-vba/Rubberduck] build for commit a401e52f on unknown branch: 63.75% (target 0%)

@Comintern yes - there's another issue where we need to ship with trace level as the default. After a first successful startup, change to the original default of none (?)

11:06 PM

Yeah, or the first time it exits normally.

We had few incidents where they got a dud that wouldn't even load and they had no way of interactively changing the log level. Not exactly the best user experience.

[rubberduck-vba/Rubberduck] codecov[bot] edited comment on pull request #4446: Add Funnel.png to resource project as Resource

> # [Codecov](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/4446?src=pr&el=h1) Report
> Merging [#4446](https://codecov.io/gh/rubberduck-vba/Rubberduck/pull/4446?src=pr&el=desc) into [next](https://codecov.io/gh/rubberduck-vba/Rubberduck/commit/de6f99412db1492eb938dd94fa9ce43044716e5a?src=pr&el=desc) will **decrease** coverage by `<.01%`.
> The diff coverage is `n/a`.

```diff
@@ Coverage Diff @@
## next #4446 +/- ##
==========================

Hmm. That's a thought. I think I like that idea better. That would cover the situation where startup was OK but something else went wrong after the startup.

We'd need to track whether or not it was explicitly changed in the settings dialog though.

probably maintain a hidden property in the settings file

like <UserChangedLogLevel>false</UserChangedLogLevel>

11:12 PM

Yeah, not hard by any stretch of the imagination. We do the same thing for prompting about legacy Smart Indenter settings.

11:40 PM

I think I've become comfortable with the idea of events. Now to start using them.

« first day (1594 days earlier) ← previous day next day → last day (1586 days later) »

Transcript for

Oct '1817

VBA Rubberducking

This chat has moved to Discord: discord.gg/MYX9RECenJ

antlr com-interop parsing rubberduck vb6 vba whiskey-tango-foxtrot

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile