« first day (2513 days earlier)      last day (150 days later) » 
00:00 - 13:0013:00 - 00:00

12:02 AM
[rubberduck-vba/Rubberduck] 1 opened issue. 4 issue comments.
[Zomis/FactorioMods] 4 commits. 1278 additions. 408 deletions.
[Minesweeper] Games Played: 92, Bombs Used: 66, Moves Performed: 12727, New Users: 9
12:24 AM
            var results = usages.Select(usage =>
                new SearchResultItem(usage.ParentNonScoping,
                    new NavigateCodeEventArgs(usage.QualifiedModuleName, usage.Selection),
                    GetModuleLine(usage.QualifiedModuleName, usage.Selection.StartLine),
                    new Selection(1, reference.Selection.StartColumn, 1, reference.Selection.EndColumn).ToZeroBased()));
I think this is going to work :)
12:54 AM
omg it's full of stars off-by-ones
1:08 AM
argh wtf I was committing just one file.. so apparently that went into a commit called "fix NRE"
amends commit message
1:28 AM
@MathieuGuindon does everyone use VS in dark mode?
or more specifically with a background color that would be described as that
nah dark mode/theme is exactly it -- I think more people than we think use a light theme
oh interesting
i asked because i think the majority of ide depictions i have seen are of dark theme
...they're probably still a minority though :)
people like a dark theme I guess
1:32 AM
why do they like that I mean
is it easier on the eyes or something?
or why do you like it I guess
actually nvm trying it out myself i see it is basically the same effect as when monitors/peripheral default color went from light to dark
that's a good way to find out!
@MathieuGuindon yeah it is
i should always assume I can guess something from trying myself before asking I suppose
anyways I have a couple actual questions but assume you are still trying to finish what you were working on so I will stop bothering you until you get that done
no worries, just post them anyway -- I'll ping you with an answer when I can, or you'll spark a discussion with someone else lurking :)
hey @Hosch250
2:04 AM
RE: Anyone can contribute - then you get info vampires like me trying to learn.
that's... actually a better thing than you're making it sound lol
Mug set the bar on the ground for me to come in and contribute.
Self-deprecating humor.
I had some experience. Just no idea where to learn more and properly ask questions.
A bit more reading docs for Moq and getting my hands dirty with examples should finally put me in a spot to actually understand mocking.
2:35 AM
with the upcoming release, the mocking API is going to ship as an experimental broken feature that can still be played with; there will be blog posts about using it :)
@IvenBach wait, you never were mocked as a kid?!? What kind of school did you go?
@MathieuGuindon FWIW I'm planning to just build the .NET types by hand. It's too flaky to rely on the .NET FX's autogenerated types
that moq wrapper is just mind-blowing awesome, you're building literal magic into RD there
I'd rather that it was more stable. :)
a nice extra, yeah
hopefully I will have more time. my evenings this week has been consumed by extra work. :(
2:45 AM
no rush - I meant after the upcoming release, not with
the SearchResult highlight will be in that PR
..I'll cherry-pick it into the find all references PR
okay questions are related to css:
1. is using display: grid always the correct way to align things in a pseudo table
that, ...I've no idea lol
2. why are buttons being forced to a minimum width (i set width of 6px and they are much more)
2:59 AM
are you using a framework? e.g. bootstrap?
your best friend is F12/devtools here - the button probably has a padding or margin defined
@MathieuGuindon bootstrap comes prepackaged with the default blazor serverside project template but I discarded it because it was overwhelming me
there is still some predefined code in site.css however so maybe that is doing it
i did actually use dev tools to figure out something else but did not think of it for the buttons because of course
and I'll reask the first: is it valid to just use a set of position: <whatever> attributes to achieve that layout?
I've no idea, I just google whatever I want to do and mess around until I get something that looks somewhat as intended
okay so there is no best practices wrt that
but learning just bootstrap makes things much easier though
@MathieuGuindon this was basically the approach i had been taking so far lol
@MathieuGuindon why?
3:06 AM
the first rule of css: you do not talk about css
get to achieve things faster, and then other frameworks do things differently of course but the concepts generally transpose I think
the second rule of css: you googlewing it
@MathieuGuindon it felt like anything it did I could do myself? I can see it does a lot but it was so overwhelming to look through to tweak. 10k lines or something ridiculous like that.
@theVBE-it'srightforme anything any framework does you could do yourself
3:09 AM
it's solved problems already
I guess it felt like I would learn CSS better if I defined everything myself as well
definitely fool around and break stuff
I did not actually delete the included bootstrap, I just moved it to another file
and have used as reference a couple times even
"break stuff" seems to be the third rule of css based on my experiences so far
it is a little unintuitive at the start compared to like a userform
I could be very wrong, but I see it a bit like XAML styling: you could redefine exactly how any part of any UI element gets to look like -- but then that's a lot of work, so you use a decent-looking framework, and then tweak it to your liking.
yeah not too far from XAML, but then XAML is worlds apart from a UserForm already
okay can button size be forced by some setting regarding padding/margins around the font
@MathieuGuindon this is the sense I had from the ~10 seconds I spent looking at XAML
i mean it felt very off compared to a form as well
I instantly understood the dislike @this expressed for WPF
3:17 AM
@this Surprisingly I wasn't.
Mocking in VBA? That sounds like it'd really make a difference.
love this
Oooh. Highlighting the results?
and it highlights Me as a reference to the Ship class
That's nice.
3:21 AM
@MathieuGuindon the tweaking it part was the problem with bootstrap though
it did not feel easy to do that because so much to sift through
without highlighting, a line like If Me Is Ship Then would come up twice - one takes you to the Me, the other takes you to the Ship, but they'd both look identical save for the selection
@theVBE-it'srightforme did you read the tutorials?
@MathieuGuindon hearty guffaw
Nice! FWIW, maybe it's just the screenshot but I think the yellow is a bit too light
I am reading now
@theVBE-it'srightforme are you aware you can edit css directly in the devtools?
3:24 AM
@this maybe blue with white foreground would be better?
tbh I spent more time googling "what is bootstrap" and then "do I have to use bootstrap" and found enough people saying they didnt use it and didnt like it for people learning to justify dumping it
I've found that the easier way to experiment without actually editing hte file
@this insta star
Possibly yes, as long it's easy to read and still pop out
i will try that
actually amazing ty
nice to get instant feedback on changes
3:26 AM
you're welcome! That helps a lot in experimenting.
Too bad I can't say that about XAML...
and perfecting things finely too I would imagine
@this actually laughed out loud at that
okay so using that I see that <something> is forcing button height to be at least 8px regardless of the font size
Bootstrap is awesome.
it does not tell me any rule is being applied to the button except the one I defined the apparently too small height
Standard styling across the site is great (if you don't use a framework, you won't have it--web dev here), but the best is the reactive layout.
same with width and 22px
reactive layout?
3:33 AM
The utility classes are awesome too.
Resizes for different screen sizes.
And the docs are easy to read for once, unlike most.
That and lodash has the best docs ever.
resizing for different screen sizes is compelling
I'd offer to help you learn bootstrap, but honestly I'm not around enough to be very helpful.
@theVBE-it'srightforme in the same css window you should be able to see the inheritance... i'm sure you already saw that some changes you may do get struck out indicating it's being overriden
But if you don't understand the docs, do feel free to ping me. I usually check here once or twice a day.
@theVBE-it'srightforme FWIW rubberduckvba.com is built with bootstrap
3:37 AM
@Hosch250 will do ty
It's super easy to use. Start with the grid system and button classes, and you'll know most of the important stuff.
is there an easier way to import it into a project than the 10k loc that I got with the default blazor serverside project template?
Skim the list of controls and utilities to see what's available, and check it as you need it.
Add it to the head of the page from a cdn.
If you don't want to self host it, which is just what Blazor does.
But, even if you self host, you never touch or read that file. Just put it in an out of the way location.
Just like a nuget package.
Anyway, nearly 11pm here, and I'm already in bed, so goodnight.
@this I like Yellow tbh... is bold-italics too much though?
@this there is nothing indicating interaction with buttons
i tried even setting min-width: 1px and using !important tags on both width and min-width
@Hosch250 gn ty for input :)
3:42 AM
I cannot not read !important as anything other than "not important"
that works for me thought I htink just bold is enough
if there's nothing struck out, then the clue is usually with the parent element, I think. (winging here!)
going with that
That works!
I htink they'll appreciate that little things. Nobody cares about compiler warnings being fixed. Everyone cares about a prettier ducky. :D
honestly I kinda liked the bold italics
only because the bold appears too similar to the section header
either is probably fine realistically
@this completely true
@theVBE-it'srightforme that's a good point
3:49 AM
Why not make it configurable?
worry about that when we get themes.
for now, we do it the Apple's way.
                var highlightRun = new Run(escapedXml.Substring(highlight.StartColumn, highlight.EndColumn - highlight.StartColumn + 1))
                    Background = Brushes.Yellow,
                    Foreground = Brushes.Blue,
                    FontWeight = FontWeights.Bold
injecting a dependency into a xaml converter is going to be fun :)
@this if you don't like how we do it you suck and should feel bad about yourself?
3:52 AM
pretty much, LOL
@theVBE-it'srightforme at the price they get their ducky for, they have no choice but to like it!
it's probably just me but I'm also more inclined to do it Apple's way mainly because I find that the more configurable it is, the more messy it ends up.
What is Apple's way?
also the user overload:value ratio not necessarily great
3:54 AM
> Here you go. Take it as we give it as that's the only option you get.
@IvenBach that way or the highway
^^ ??
@IvenBach see my earlier message
(so yes)
nothing I try seems to resolve this issue of button width :/
i can work with minimum height of 8 but 22px is way too wide
3:56 AM
8px buttons seems rather small
22px is just enough to fit at 16x16 icon and have a bit of a margin
just wondering - is it set like block style where it is forced to take up 100% of the containing element?
it is meant to be small, it is just to remove an item from a list
just has a plain text x as label
@this potentially?
@MathieuGuindon how large is the tab close button in the RD references results window above?
that's a 16x16 icon
tucks tail in shame
blue instead of black looks good to me I think
4:02 AM
hmm i might be blind but looks black to me
I have decided to read the MDN article on the button element btw
ah I see it. after I zoomed in
do not like it at all tho
agree. I think black was more legible
yeah definitely
ok bold black on yellow it is
4:05 AM
See, Mr. Jobs had it right. Less is more. :)
hopes that Mat doesn't put in blue to spite for that comment
@this the less you respect the customer the more they will buy your product!
LOL that's eerily true.
You know, you just put Steve Jobs side by side with Ozzy Osburne. Same energy.
expressed very very differently but I see it
@MathieuGuindon can you have like half bold font?
that would maybe differentiate from headers while saving busyness of italics
4:14 AM
> The feature had been broken for some time; this PR restores its functionality.

Also introduces the possibility to highlight (bold, black on yellow) a `Selection` in the search result text of a `SearchResultItem`, and leverages this possibility for _find all references_.
@theVBE-it'srightforme SemiBold incoming...
@MathieuGuindon hey i legit thought that was a thing...
@theVBE-it'srightforme it is:
^ semibold
@this thoughts? I like it but am also biased because I suggested it.
it is actually hard to tell the difference with regular bold tbh
actually no it is not
4:21 AM
looks a tad lighter but is fine by me.
I like it
"Semibold ought to be enough for anybody."
well I am glad I suggested it then
@this does anything look weird/off to you in the argument-aware PR?
4:24 AM
Again, nicely done, @MathieuGuindon! That'll be a crowd-pleaser.
apparently just had to set padding: 0
guess who needs to read about padding more!
oh you requested my review. Weird.
@this thanks!
4:26 AM
@IvenBach btw before I forget to say this I think it is being unfair to call yourself an info vampire
@theVBE-it'srightforme Self-deprecation runs deep with me.
hm it looks better aligned in the designer
@IvenBach I know and it tends to with me as well but I find that too much of it can be counterproductive.
Fine line between staying grounded and beating yourself down I mean.
Also you have helped me several times recently so it is not technically true anyways. :p
okay so css box generally is content + padding + border + margin
4:38 AM
I think the search results toolwindow needs a toolbar
content is obvious, padding separates that from border, border separates from margin, margin is the "force field" so to speak that isolates the box from others
@theVBE-it'srightforme I contribute as best I can. Hope to keep helping for a long while.
I’m overdue for my own blog post for how to use RD with Git.
@IvenBach feel free to add expand/collapse buttons to a new toolbar for the search results toolwindow ;-)
[rubberduck-vba/Rubberduck] build for commit b924ce36 on unknown branch: AppVeyor build succeeded
do you mean a toolbar within that window?
4:39 AM
yeah, like the other dockables
@MathieuGuindon just to double check -- you are adding additional references here but only if hte declaration is a parameter declaration. Does that it mean if I find all references on the object rather than the parameter, it won't include the parameter usages?
fires up Excel for visualization help
@MathieuGuindon want to open an issue for it and assign me? I’ll forget if not.
i have made like no progress on my actual work but have felt very productive over the last 1.5 hours
@theVBE-it'srightforme that's 90 minutes well spent, that will make it much easier to actually progress :)
4:42 AM
i suppose that is true
and I contributed a slight bit to RD too!
on that note, what would be an example of a dockable toolwindow
the code explorer
the immediate pane
going to use a fresh wb so I know RD will actually parse
@this "on the object rather than the parameter", meaning looking for the references to the type of a parameter?
oh okay i see toolbar in code explorer
4:45 AM
that would be correct
@IvenBach blog post on how to use Git in general would be helpful even (I mean for beginners)
well it'll find that parameter's As clause
as long the result are the same from either location, then that's fine - it's just a bit confusing to see a special case in only one direction, that's all
hm wait no, no if you're on an argument it links to the parameter, and if you're on the parameter it links to both local usages and argument/call sites
4:48 AM
@IvenBach well written from what I can tell :)
I like the spanish style question punctuation in the second paragraph as well
@MathieuGuindon that seems correct?
let's use an example:
Public Sub doh(msg As String)
End Sub

Public Sub derp()
    doh "hi"
End Sub
if the caret is on msg, and I Find all References, will it pick up the "hi" literal?
or only when I have the caret on the "hi" literal?
OK, in your PR you had the screenshot demonstrating the latter but not the former, so wanted to be sure that worked both ways
this one?
4:55 AM
no, this:
in this example the caret is on the literal "Done!"
but MsgBox is not a user-defined function so we can't put a caret on the Prompt parameter.
also that was before the commit that lit up the references command in the toolbar
Were it, though, I would expect that Find All References would still show "Done!" argument as a reference even when the caret is on the Prompt parameter.
If that's the case already, great! Just not clear from that code I linked since it seems to deal with ParameterDeclaration as a special case
hold on, checking out the branch - I'll confirm this
will go ahead and put in a comment on that so you can then address once confirmed
that's correct: when a parameter declaration is selected we add the argument refs
that's why it's special-cased
and with the caret on an argument ref, the target is the parameter declaration
5:00 AM
gotcha, so this is handling both sides of the equation equally. Very good.
wait, that's the same case as the screenshot I referenced. I was talking about the caret on the parameter, not on the argument.
in that case, it needs to find the arguments as references, right?
^ caret was on msg
ok, good
(I admit clicking that button made that non-obvious lol)
^ that was the only thing I could find beside the question I asked above. :)
@MathieuGuindon yeah I was about to ask but you clarified, lol
5:04 AM
PR does need more tests though
@Duga I don't know why that is the case
@this thanks for eyeballing/reviewing!
css feels like it will be pulling teeth into some moment where I just "get" it and it becomes effortless
moreso than things generally are I mean
sure then one month later it kicks you in teeth and then drop kick you out of pure spite.
Advice: bring a cup.
maybe a helmet, too.
scribbles furiously
though with that said you could probably also generalize your statement to any language/tool used within the realm of software development
5:20 AM
based on git branch -r --merged any reason we're keep those branches which have already been merged into next?
patching, maybe
otherwise no, they're safe to delete
@this thoughts on moving Tokens.cs from Rubberduck.Parsing to Rubberduck.Resources?
I was just about to do it and then I saw codelens 99+ uses and went looking for a move command that wasn't there
[rubberduck-vba/Rubberduck] IvenBach deleted branch retailcoder-patch-2
[rubberduck-vba/Rubberduck] IvenBach deleted branch retailcoder-patch-1
[rubberduck-vba/Rubberduck] IvenBach deleted branch patch-iss
[rubberduck-vba/Rubberduck] IvenBach deleted branch InspectionTypeDefaults
[rubberduck-vba/Rubberduck] IvenBach deleted branch RubberduckAPI
[rubberduck-vba/Rubberduck] build for commit 47499bdc on unknown branch: AppVeyor build succeeded
5:37 AM
@MathieuGuindon aren’t they generated by antlr?
funny, R#
@this no, it's our static class
oh. feels redundant since wedefine it in grammar
but in that case yeah resources would be a better place
Antlr tokens are int though
6:11 AM
now that it is time for me to sit down and focus on tinkering with the css my steam library grows ever more appealing :/
wait I thought you could inline variable declarations in if blocks
Get some sleep Mug.
6:34 AM
@IvenBach do you remember which c# operator only evaluates the right side if the left side returns true
oh lol it is the simple && isnt it? maybe I should also listen to your advice :/
6:58 AM
C# short circuits meaning if (a && b) where a evaluates to a false then it will not proceed to evaluate b because with the left operand being false the && can never be true, regardless of what the right operand is.
if (a || b) is the other side. When the left operand a evaluates to true then it doesn't matter what the right operand evaluates to since true || ____ evaluates to true.
Building a truth table on each of the inputs will help illustrate how the operands will result.
7:15 AM
@IvenBach :+10:
I would accept that if it were an SO answer.
@IvenBach 'night!
7:37 AM
[rubberduck-vba/Rubberduck] build for commit 43a4d41d on unknown branch: AppVeyor build succeeded
7:57 AM
> There are multiple pieces in the codebase that need information about the qualifying reference/declaration of a reference, e.g. to determine whether the reference is a member access on the return value of a specific member. Currently, checking this is achieved by navigating the parse tree each time, also taking into account with blocks. This PR moves the qualifier resolution to the BoundExpressionVisitor, which already has all the information present to easily determine the qualifying...
reference. The following qualifier structure has been implemented: - Member accesses are qualified by the reference for their lExpression. - Member accesses return their own reference as the one qualifying a potential access. - For (recursive) default member accesses the innermost access is qualified by the reference of the lExpression and each other access is qualified by the immediately contained one. - For (recursive) default member accesses the outermost access is returned as potentially...
qualifying. - Array accessed are qualified by the reference of their lExpression (potentially with default member accesses in between), i.e. by the reference of the array variable or array returning member. - Array accesses return themselves as potentially qualifying. (To get the array, one has to go up the chain again.) - Index accesses other than array and default member accesses are qualified by the reference of their lExpression. - Index accesses other than array and default member...
accesses return the reference of their lExpression as potentially qualifying. - Dictionary access expressions work like default member access expressions. - Other expressions neither take nor return a reference as (potentially) qualifying. This PR is still WIP; it needs a lot more tests. Moreover, the places currently doing the work manually should be ported to use the qualifying reference, if possible.
@Duga That should have been a draft.
accidental submit?
No, I just forgot to hit the drop down to change it to draft PR.
oh sorry I forgot draft PR were a thing, so was only thinking in context of the text you wrote
8:23 AM
[rubberduck-vba/Rubberduck] build for commit f9273352 on unknown branch: AppVeyor build succeeded
8:40 AM
Regarding your discussion on VS theme yesterday, I use light theme.
There are actually studies showing that this is easier on the eye in well-lid surroundings.
9:01 AM
@M.Doerner The eye strains less to pick out text on a background that is the same general color as the surrounding environment?
The studies show that the eye generally easier picks up dark text on light backgrounds. The thing regarding well-lid surroundings is that it is also hard on the eye to look at a bright screen when it is attuned to a lower light level because of dark surroundings.
However, the effects do not seem to be too significant.
@MathieuGuindon I would like to suggest no background color at all. I think making the reference bold already does the trick and is less distracting.
9:26 AM
Moreover, note that semi-bold quite often actually renders as bold.
I think that is a problem caused by running the WPF control in a WinForms container.
10:10 AM
@MathieuGuindon R# has a move file refactoring. You can access it in the solution explorer's context menu.
@IvenBach I have just started to look at your blog post. Is it possible to change the spacing above headings? I think it looks a bit odd that the space between the heading and its section is larger than the space between the previous section and the heading.
10:39 AM
@MathieuGuindon Hi Mat. This is what I was talking about. (I do have a bigger project, not a simple one but this is restricting my workflow a lot)
I click between two modules. You can notice that by seeing the cursor. At the end, I press only the right arrow on the keyboard and it slows immensely, almost freezing the program.
10:57 AM
@MathieuGuindon Add a touch of marquee and I think you'll have it perfect!
@FreeMan someone is about to not be so free and not so manly.... ;-)
@Duga @IvenBach was patching the ISS? Cool! We have a rocket surgeon!
@this lol
TBH, highlighting is just fine for me. I don't think it needs any font change at all, but I'm a bit late to the party (as usual).
That makes me fashionable, right?
11:26 AM
Fashionably late. :-D
eh... close enough.
English's funny beast. You'll never hear anybody say anything like "My early husband and I were happily married for 39 years." Who decided that "late" had to be a polite euphemism for having shuffled off the mortal coil?
11:47 AM
probably the same person who decided that "having shuffled off this mortal coil" was a polite euphemism for kicked the bucket.
TBH, though, my wife might refer to me as her "early" husband after I'm gone. I despise being late for things, but being late is her hobby, and she's committed to it.
@FreeMan I LOL'd at the use of another euphemism to explain yet another euphemism. Can't just say "he died"
That was the intent. ;)
I just got the news, Bobby. Our dear uncle Fred has departed.
Where did he go to?
He is no longer with us.
I know, but where did he go?
I'm trying to say that uncle Fred has bought the farm.
Oh, can we visit his new farm?
No, no. It's not like that. He has kicked the bucket.
But I kicked a bucket yesterday. Do I have to go?
No, dear, he is pushing daisies.
Why can't he pick it up instead of pushing?
Because he's six feet under.
Under what, mom?
Dear, I'm trying to explain that your uncle Fred is now sleeping with the fish.
12:56 PM
oh goody, another mid-day reboot to install Windoze updates. Thank you corporate IT. :(
@this noice!
@FreeMan We boot because we care!
Thank you Monsters, Inc.
sorry couldn't find a rhyme for care in the context of rebooting.
We boot because we toot!
12:59 PM
Ah'll be bahk...
00:00 - 13:0013:00 - 00:00

« first day (2513 days earlier)      last day (150 days later) »