« first day (855 days earlier)      last day (3177 days later) » 

1:00 PM
(and Pops had problems with the 'é')
 
@SimonForsberg lol
 
I never expected to hit HNQ with it though.
 
I suppose that counts as a good reason :-)
 
Anyway, I'm going for a walk. Feel free to ping, will read later.
 
@Mast Ideally you shouldn't have to predict that - it should just be a nice surprise.
 
1:01 PM
Hasn't there been a meta about this yet? Perhaps we should.
 
/me confused... about what exactly?
 
BF in titles?
 
If no one complained last time I'm guessing you don't need a meta post for Code Review - once HNQ is fixed no one here is going to complain so it doesn't seem important
 
yaaay, back to then
 
Yes, it does seem that the HNQ is the problem and not the title itself.
We won't fix the HNQ filter by editing the title of this particular question.
 
1:05 PM
If you decide not to censor it, it might be worth mentioning it on that MSE post as an example so they can speed up the fix...
 
@trichoplax Shog9 has already been informed about it in the moderator chatroom. Because of today being Sunday though, I don't expect an answer about it quickly.
Meanwhile, this is an interesting read:
The Scunthorpe problem is the blocking of e-mails, forum posts or search results by a spam filter or search engine because their text contains a string of letters that are shared with an obscene word. While computers can easily identify strings of text within a document, broad blocking rules may result in false positives, causing innocent phrases to be blocked. == Origin and history == The problem was named after an incident in 1996 in which AOL's profanity filter prevented residents of the town of Scunthorpe, Lincolnshire, England, from creating accounts with AOL, because the town's name contains...
 
@SimonForsberg Thanks - I had heard of that, it's an interesting problem. Since there aren't a long list of languages containing this word I'm guessing it's just a matter of adding bf to the list of blocked words
I agree you wouldn't want to block partial words - that could be disasterous :)
 
1:28 PM
Appreciate the clarification. I have taken all of your advice and written three functions that I hope you would keep under consideration reviewing, if and when you have time: codereview.stackexchange.com/questions/103954/…kexxcream 17 secs ago
 
Zak
0
Q: Visiting nearby attractions

Aman AroraI had a problem to solve for a challenge.Suppose I visit a city and stay at a hotel. I am given the latitude and longitude of the hotel location. I am also given the latitudes and longitudes of the nearby places of attraction in the city.I had to calculate the distance between the hotel and each ...

Also seems like broken code
 
0
Q: Custom error/fatal error and exception handler in PHP

kexxcreamIntroduction I wrote three functions that aims to log errors and exceptions as well as show a user a generic error page. Would appreciate some feedback on it with regards to the PSR standards and if I have left out anything of value. Memory issue: I am aware that the memory issue has not been...

0
Q: Visiting nearby attractions

Aman AroraI had a problem to solve for a challenge.Suppose I visit a city and stay at a hotel. I am given the latitude and longitude of the hotel location. I am also given the latitudes and longitudes of the nearby places of attraction in the city.I had to calculate the distance between the hotel and each ...

 
@CaptainObvious slightly overcommented...
also does not include a fix for when it runs out of memory, but OP is aware...
hmm..
 
I think we have reason enough to put it on hold. OP is free to clarify the question. I don't see an indication that he's looking for a review.
 
Good morning everyone :)
 
1:37 PM
morning Morwenn
 
I've reworded the part about the known issue, as asking the CR community for help fixing specific programming issues is out of scope (that's SO's territory ;-) — Mat's Mug ♦ 22 secs ago
 
+195 rep today, 5 away from capping for the first time ^^
 
hears sleighbells
 
No rep for days here.
 
@Mat'sMug Of course, was not my intention. Thanks for the edit. — kexxcream 36 secs ago
 
1:40 PM
On the other hand...
 
I may soon be ready for my first question :D
 
Thanks Santa!
 
I'll post proof when caching catches up.
 
@Morwenn last answer 08/28 ...slacker!
 
1:42 PM
@Mat'sMug Hey, I've been working on my vergesort meanwhile :o
It's like TimSort except it's actually better.
 
TimSort is also buggy in Java....
 
in VBA Rubberducking, 14 hours ago, by Duga
[rubberduck-vba/Rubberduck] 76 commits. 1 closed issue. 1 issue comment. 37469 additions. 33526 deletions.
 
holy carp.
 
no wonder the project seemed dead for a month.. about 1/3 of the code base has been refactored
 
user image
4
 
1:43 PM
congrats @Mast!
 
congrats @Mast!
 
Thank you :-)
 
@Mast you got that badge before me now..
darnit
 
@Vogel612 wait, you never capped?
 
Mortarboard on a Sunday
3
 
1:45 PM
@Mat'sMug nope. not here, not even on my first day..
I capped twice over at MSE though..
 
@Vogel612 You hit 185 once here, I think that's the highest you got.
 
for that rant about how the new graduation sucks :D
 
chillworld got over 300 score on one of his MSE questions, wonder how often he capped on that alone
Oh, MSO, not MSE
303
Q: Be careful when recommending Code Review to askers

chillworldCode Review is becoming well-known on Stack Overflow. The folks over on Code Review are happy about that. The only problem is that there are a lot of questions redirected to CR which don't belong there. So I want to ask that the users who direct askers to CR (or request migration of questions...

 
@Mast 278 is my max actually..
 
new BF question incoming
 
1:49 PM
1
Q: Mini BF interpreter

user83893Many Brainfuck implementations I've come across generally are obtuse, verbose or over-the-top. But I've always seen Brainfuck as "easy to implement", especially for beginners. My original need for this was to be able to have an actual Brainfuck "machine" that executed instructions to aid with co...

 
Seems decent.
 
2:02 PM
Ok, so my sorting algorithm seems to always be faster than TimSort and a bit slower than a pattern-defeating quicksort at worse.
Sounds robust enough.
 
Monking
 
And it's not even totally optimized.
 
I have no clue how to tell whether the algorithm is O(n log n) or not, but since it beats or is equal to the best sorting algorithms, I guess that it runs in O(n log n).
 
saw that
all right, RL calls. and I just posted what should be my 100th qualifying question for [badge:socratic]
 
2:11 PM
@Mat'sMug I thought I quoted the wrong comment, then realised that I quoted the right one - but it was too late to undelete
 
@Morwenn just so happens that monkey and I created a tool to figure such things out :) Only runs on Java code though...
 
@SimonForsberg Well, port it to Java then :p
 
My mind goes to mush when you talk about O(n log n)
I know what it is, but I have 0 formal education in algorithms
All i know is, generally, nested for is bad (O(N^2))
 
later CRitters!
 
2:14 PM
Later :)
 
0
Q: Sorting Dates (DD/MM/YYYY) with Insertion Sort

JekasGI'm having troubles with a coding problem. Yes I need to write a C++ program which sorts dates. I've tried several methods. Using 2D arrays and now parallel arrays. The problem I have with my parallel array to keeping tract of those which haven't been sorted yet. That is the YYYY has been sorted...

1
Q: CommandBars, Buttons and Commands: Take 3

Mat's MugFollowing up on CommandBars, Buttons and Commands: Take 2, I've refactored quite a lot of the Rubberduck code base (like, a whole third of the entire code base has been affected), and successfully untangled the RubberduckMenu and gave each command its own place to be. So, "Shared" commands that ...

 
@CaptainObvious Congrats on Socratic ^^
 
@Mast not quite yet
 
@Vogel612 He was at 99, this is 100
 
Any time now SE...
SE is mocking us.
 
I guess they don't update this stat too often
 
probably better to post at codereview.stackexchange.comachampion 38 secs ago
 
2:39 PM
@Mat'sMug It takes caching forever to catch up, but congrats on your Socratic :-)
 
2:50 PM
Thanks!
 
it's really taking long eh....
 
Morning everyone!
 
hey Ethan
 
How's everyone doing?
 
Mat'sMug became the 2nd person to recieve Socratic
4
 
2:53 PM
Nice
 
wohoo, congrats @Mat'sMug
@Mast Thanks for posting that Brainfuck question, I have wanted to add some analyze-like features to my Brainfuck interpreter. Your question gave me a reason to do so.
 
Please note, if you do choose to post this to Code Review, the title must describe the purpose of the code, and the body should elaborate on that purpose a bit. — Ethan Bierlein 40 secs ago
[Vogel612/DynamicForm] Vogel612 created repository
 
@Duga @EthanBierlein saying "must" is a bit of an exaggeration. When guiding SO users to CR, the title isn't that important really. We can deal with the title once they get here and once they include description about what the code does.
 
@SimonForsberg Yup. I actually meant to say should, but yeah.
 
3:14 PM
Greetings
 
Greetings @IsmaelMiguel
 
How's it going here?
 
That's always good to know
Do you use StackExchange's app on Android?
 
@SimonForsberg My pleasure.
 
3:33 PM
Monking
 
Hi.
 
Greetings
 
Nomking
eating
 
It's about time I come up with another blog topic.
 
Generics?
 
3:38 PM
@nhgrif oooh, any ideas?
 
Generics and inheritance can be extremely difficult to do right for certain patterns.
 
Perhaps the benefits of composition over inheritance?
 
See Eric Lippert's Wizards and Warriors series, for example.
 
My blog topics are typically slightly more generic than those sorts of topics.
I was thinking about writing about unit testing and its value.
 
That's good too.
 
3:40 PM
Always a good topic.
God knows I preach enough about it at work.
 
In a very general sense (what's the point of unit testing, what does it accomplish, what does it not accomplish, and how to convince someone who doesn't believe in it of its value... specifically non-programmers)
 
@nhgrif too generic for generics
5
 
But at the same time...
It's September, and I feel I should right something more September-appropriate.
This is an example of a September-appropriate blog entry.
 
Maybe a don't do this guide
'Schools out; you've decided that because you can install .exe files you're good with computers and that you'll start making apps'
 
Maybe I should tell students about how many things their programming professors will be wrong about... and the types of things that they should endeavor to learn while still in school (but that schools, as a general rule, do not teach)
 
3:43 PM
You've done comments already, haven't you?
How about spacing?
 
@nhgrif how specific is that? language-related or just in general?
 
And naming?
@Quill Presumably in general.
 
Like version control, dependency management systems (like Cocoapods for iOS/OS X development... I'm sure there are others for other platforms), using 3rd party libraries in general, unit testing
Just a light dusting of each topic (some of which might point back to previously written articles, and some of which serve as bookmarks for topics I need to discuss in more detail in the future)
 
That @nhgrif. Please. There are so many professors out there teaching bad habits.
 
Everything that anyone in the guild of programming needs to know.
And that no one out of the guild knows.
 
3:45 PM
The secret handshakes of the programming guild
5
 
Hmm... I feel like I should write a open source "free as in puppy" piece.
Many people don't consider the maintenance cost/risk of using an open source library.
 
Versus the maintenance cost/risk of writing it yourself?
 
Umm, why in the world does this have three upvotes?
@nhgrif I think he means the risk of paying someone else to write it for you.
There is still the risk of bugs, but there is someone who's job it is to fix it.
 
@Hosch250 No clue, it wasn't really funny, I must admit. Perhaps the lurkers want to torment us later with my bad jokes?
 
3
Q: Sorting dates (DD/MM/YYYY) with insertion sort

JekasGI'm having troubles with a coding problem. Yes I need to write a C++ program which sorts dates. I've tried several methods. Using 2D arrays and now parallel arrays. The problem I have with my parallel array to keeping tract of those which haven't been sorted yet. That is the YYYY has been sorted...

@Quill Not that comment, I forgot to post the link again.
It sounds like he is wanting help debugging/adding new features to his code.
 
3:50 PM
@Hosch250 Oh, I see. The message still applies either way
 
@Hosch250 Wow, I have completely misread that one. It looked okay when I reviewed it.
I read it like it was already sorting. I guess from the consternation it isn't?
 
yeah, I also misread it, and I've edited it, so you should be able to Downvote/Unvote it
 
Does anyone here have a good source control tutorial blog entry and would like to be linked to by my blog? I've started writing this entry... I don't want to get into the nitty-gritty of all of these aspects... I just want to compile a list of things that students basically have to go teach themselves since universities tend not to teach these things
 
No @nhgrif. Yes. Kind of.
 
3:58 PM
@Quill Unvoted it, I haven't tested the code (yet)
@nhgrif I'll pull you a shortlist I use which is on SO
784
A: Import existing source code to Github

PeteIf you've got local source code you want to add to a new remote new git repository without 'cloning' the remote first, do the following (I often do this - you create your remote empty repository in bitbucket/github, then push up your source) Create the remote repository, and get the URL such as...

Very basic, very good.
 
I think a lot of people think "hey, great, there's this library", but never stop to think that they might have to maintain it themselves.
You always run the danger of ending up with your own fork.
 
@Mast I'm looking for something a bit different from this. Assume the repo is already set up (before you were hired) and you know absolutely nothing about source control. You need a very basic tutorial for checking out, branching, checking in, merging, etc., etc., etc...
 
I'm not saying it's always better to just write it yourself, but there are situations where you might. I think that gets overlooked.
 
There's two sides to the argument.
 
That's awful dependent on the source control system. TFS is very different from Git for instance.
 
4:02 PM
I'd rather err on the side of using a library when I should've written it myself than writing it myself when I should have used a library.
 
Agreed.
 
The latter results in dinosaurs that are 20 years into their career and wrote everything from scratch and have a long list of justifications for why their case is a special snowflake.
 
I've just got myself wondering when/how I would justify DIY
 
AND, when you have new people come on to the project, there's a chance they're perfectly familiar with that popular 3rd party library from Github, but there's not a chance they are familiar with your library.
 
That goes for any code base though.
 
4:04 PM
I'm in a project where... effectively everything in the project was a DIY.
 
Really not the point I was getting at though.
 
I know.
But it needs to be part of the decision making process.
 
Yes. It does.
 
I think you're far more likely to run yourself into extra work if you are DIYing things too much.
 
I should write a Git tutorial.
 
4:05 PM
@nhgrif I'd say you should first try to set one up yourself and learn from that.
 
For beginners.
 
I've just seen a lot of "just find some FLOSS, it'll save time" lately. Which is true, but people aren't considering that we might end up maintaining our own fork.
 
When you got those basic steps down, proceed to 'existing' repo's and how to work as a teammember in them.
 
I can name about 3 or 4 extraordinarily popular 3rd party open source libraries from github that my current project should be using but isn't because the people that wrote it thought everything they were doing is a special snowflake
6
 
Like Moq?
 
4:06 PM
And even before anyone even asks the decision-makers why that's not happening, they start explaining all the reasons why their code is so special
 
Or IoC containers.
Yeah. That's not good either.
I'm particularly thinking of a survey engine we're looking at.
 
We're not using AFNetworking, a github project with nearly 20,000 stars...
 
At least I didn't write my own C# compiler.
 
Right. Those projects are stable and widely used. I'm thinking of more obscure projects.
But that's definitely a distinction that should be made.
 
Right.
But even if you do have to maintain your own branch of a library, that might be a better starting point then scratch.
Depending on what it is
 
4:08 PM
Definitely.
 
And it might not be. That ^^
It needs to be evaluated. You need to review the source and be sure that you could fork it if you had to.
 
Or, maybe the existing option is so bad it would be better to start from scratch.
(Not too often, but possible.)
 
Exactly my thoughts. You have to look under the hood.
 
And whatever you do, if you're writing library-esque code, even if it's purely internal, it deserves to be extraordinarily well-documented as if it were being provided as an external library.
 
Anyway. RL calls. Thanks for the sounding board guys. Looking forward to your next post @nhgrif.
 
4:10 PM
@nhgrif You could write an article on that alone.
 
Untaught Skill #1: Source Control
Untaught Skill #2: Unit Testing
Untaught Skill #3: Programming Concepts, Not Language Syntax
Untaught Skill #4: The Debugger
that's what I have so far
 
What about soft skills? Like, giving/receiving code reviews.
 
Is using reasonable naming under #3?
 
Documentation
Time management.
 
I'm not particularly interested in soft-skills. I want to focus more on things that absolutely should be taught in actual programming courses but are not.
And documentation probably falls in that category
 
4:16 PM
So does naming.
And formatting.
 
Naming is part of documentation
 
OK.
 
And formatting isn't big-picture enough. Also, I've at least heard of some professors partially grading on formatting
 
Code structure. (Like, the source code file structure).
 
Project Structure
 
4:18 PM
What about concepts, like DRY?
 
I've seen lots of Qs where students are encouraged to put all their code in a single file.
 
lol
 
I understand why, but it teaches bad habits.
 
@RubberDuck shivers
 
Ikr
 
4:19 PM
I don't understand why.
What's the justification?
 
That is one thing Java got right - one class per file.
 
Easier to submit for grading.
 
That's the worst reason I've ever heard.
You know how students should submit for grading?
 
At least, that's the theory. I think it's misguided.
 
Source control commit.
Because then it's easier for the teacher to check out for grading.
 
4:20 PM
Zip file is what I had to use.
 
And I think you should say so in your post.
 
And the students are learning from day one how to use source control.
 
The prof just had to unzip to run.
 
Better than a commit, a tag.
 
That's the only way that programming homework should ever be turned in.
 
4:21 PM
@Hosch250 Naming is difficult.
 
How about the prof sets up a branch for each student?
The student makes a PR.
 
How about the professor does nothing
 
That ^
 
Indeed. Let the students do the work, like they're expected to.
 
Except on day 1 teaches the student how to use source control to turn in their homework
If you can't use source control to turn in your homework, you don't need to pass any programming classes.
 
4:22 PM
Yes please.
 
Let the prof do the prof stuff instead of handing out forks.
 
Here's an idea, programming taught by professionals instead of professors.
Apprenticeships
 
The university should have a private repository set up somehow, either through their public website that only students/professors can access, or on the university's local network
 
@RubberDuck Around here, you need a MSc to teach a BSc. How many programmers do you know with a MSc interested in teaching?
 
@Mast Around here, you usually need a PhD.
 
4:23 PM
You don't need a college degree to be a programmer.
 
@RubberDuck that system exists here in Germany..
It's a failure
 
That's my point @Mast. Those people don't know how to teach these kids what they actually need to know.
 
I went through and I learned almost everything from SO / CR
 
@Vogel612 They want to try it here as well. I'm already laughing.
 
Really @Vogel612?
 
4:24 PM
I had 2.5 years of "traineeship"
 
I'm surprised by that. Apprenticeships have worked well for millennia.
 
the idea is you work at a company most of the time, but additionally visit a dedicated school
usually it takes 3 years
the school is garbage for the programming-related things
 
Yeah. That's not what I mean really.
 
oh you mean mentorship
 
0
Q: i write a calculator code for android, how i can make it more efficient this further

Sohaib Razai am learing android, this is my first effort to write a android application , please review it and point me some mistake or improvements. Here is my github repo link calculator code some of its part is protected void number(long num) { // TODO Auto-generated method stub val = num...

0
Q: Invoke callable object passing arguments in "groups of N" using C++17 fold expressions

Vittorio RomeoInteractive Wandbox example Inspired by this StackOveflow question, I've implemented a function that, when called with a desired arity, a callable object, and a variadic amount of arguments, calls the passed callable object forwarding the passed arguments in groups of the desired arity. Examp...

 
4:25 PM
^ an example of the problem with academics & programming
 
I guess so @Vogel612. I think maybe we have slightly different ideas of what the words mean.
 
That video is roughly "Haskell is better than Java because function programming is better than imperative programming. Also, I can't even really write a Java for loop that does what I say it does. And I'm saying this thing is global regardless of context, which isn't really true."
3
I like Computerphile. I do not like that guy (that's the only video with him I've seen).
 
Agreed
 
I'm having a DRY problem here, mind if I interrupt?
private void AnalyzeCSharpSymbol(SyntaxNodeAnalysisContext context)
{
    var attributeExpression = context.Node as Microsoft.CodeAnalysis.CSharp.Syntax.AttributeSyntax;

    if (attributeExpression?.ArgumentList == null || attributeExpression.ArgumentList.Arguments.Any())
    {
        return;
    }

    context.ReportDiagnostic(Diagnostic.Create(Rule, attributeExpression.GetLocation()));
}

private void AnalyzeVisualBasicSymbol(SyntaxNodeAnalysisContext context)
{
    var attributeExpression = context.Node as Microsoft.CodeAnalysis.VisualBasic.Syntax.AttributeSyntax;
Those methods are identical except for the cast type (and that should really be a cast at the top instead of an as).
 
@Hosch250 Definitely wet.
 
4:29 PM
I could use dynamic, but that would definitely be bad.
 
private void AnalyzeSymbolForLanguage(SyntaxNodeAnalysisContext context, SomeEnumRepresentingLanguage language)
 
@nhgrif I generally like Computerphile, but I don't like that nerd either.
 
@nhgrif I can't do that.
 
@Hosch250 Can't you pass the type to be cast as an argument?
 
Your two methods can still exist, but they just call that one.
Why can't you?
 
4:30 PM
Because this is how I call them:
public override void Initialize(AnalysisContext context)
{
    context.RegisterSyntaxNodeAction(AnalyzeCSharpSymbol, Microsoft.CodeAnalysis.CSharp.SyntaxKind.Attribute);
    context.RegisterSyntaxNodeAction(AnalyzeVisualBasicSymbol, Microsoft.CodeAnalysis.VisualBasic.SyntaxKind.Attribute);
}
 
That's not calling what you just showed.
 
The signature is fixed - I can't touch that.
 
I'm not telling you to change the signature.
Your two methods are wrappers for one with the signature I just posted.
 
Oh, OK.
 
That's a pretty common technique.
It's like convenience constructors
You have one constructor with two arguments
And then ten constructors with 1 argument that call the first one with the passed in argument and a predefined second argument.
enum FooBar {
    case Foo, Bar
}

func foo(arg1: Int) -> Bool {
    return fooBar(arg1: arg1, arg2: .Foo)
}

func bar(arg1: Int) -> Bool {
    return fooBar(arg1: arg1, arg2: .Bar)
}

private func fooBar(arg1: Int, arg2: FooBar) -> Bool {
    // 90% common code, and checking arg2 for divergent areas
}
 
4:35 PM
@Mast Just for comparison, here's the run-time analysis for my FizzBuzz implementation in BF:
NEXT: 30788
PREVIOUS: 30787
WRITE: 413
ADD: 37025
SUBTRACT: 36071
WHILE: 8872
END_WHILE: 28295
Total: 172251
 
For that matter, I could just use two variables and check for null.
 
@SimonForsberg Is that a healthy mix?
 
@Mast Is Brainfuck ever healthy?
6
 
@SimonForsberg Let me rephrase that. Is that combination idiomatic?
 
Is anything idiomatic in Brainfuck?
 
4:40 PM
Well, my code isn't.
So something must be.
Law of Negation
 
I don't know what's idiomatic in Brainfuck or not, but well... when it comes to somewhat complex things (as much as you can call Fizzbuzz complex...), it tends to require a whooooole lot of loops and additions and subtractions... so, I guess it's idiomatic.
 
@nhgrif I don't see how I can do that since everything depends on the one variable.
 
@SimonForsberg Because it has a whole lot of loops, additions and subtractions, it's idiomatic. Flawless logic.
What could possibly go wrong there.
;-)
 
It gets the job done, but maybe in a year or two my Brainfuck skills will have improved so that I can find ways to improve it.
 
Regarding benchmarks, I have 2 that show significant differences between accessing a range (43.578 sec) vs memory (2.023 sec), and AutoFilter performance with 1 million rows. — paul bica 47 secs ago
 
4:43 PM
Well, at least you have an extensive analyzer now.
So you'll know whether it's 'better' or not.
 
there are other ways of making fizzbuzz in Brainfuck, I chose a quite memory-tape-focused solution, to make it extensible for other woofs and barfs etc as well.
extensive? I don't know about that. I can analyze the source code instructions and the run-time instructions being performed, that's about it.
 
It's a start
Good enough to see the difference
 
yeah, it's useful when comparing different BF codes
 
private void AnalyzeSymbol(SyntaxNodeAnalysisContext context)
{
    var csAttributeExpression = context.Node as Microsoft.CodeAnalysis.CSharp.Syntax.AttributeSyntax;
    var vbAttributeExpression = context.Node as Microsoft.CodeAnalysis.VisualBasic.Syntax.AttributeSyntax;

    // attribute must have arguments
    // if there are no parenthesis, the ArgumentList is null
    // if there are empty parenthesis, the ArgumentList is empty
    if (HasEmptyArgumentList(csAttributeExpression) && HasEmptyArgumentList(vbAttributeExpression))
That doesn't clearly show that one of the variables will always be null, so I have to use the && operator.
I guess I will just post a question on CR.
 
If you have running code and want to know how to improve it (or check whether it's best practice), head over to Code Review.SX. If you have a problem with your code, add it in your post. — Zeta 20 secs ago
 
New user killed a zombie!
2
A: Reference Implementation for Honey Badger programming language

Michael Grünewald Obviously, if you have any questions just ask. This is my first "big" project in OCaml, but I'd rather you didn't sugarcoat criticism. It looks like a very fun project, congratulations! :) I have a few comments, which all concerns the style of writing. This is an important topic, because a...

 
Nice
 
Better get it before @Jamal does: codereview.stackexchange.com/questions/103965/…
 
@Hosch250 You got too much colons
And here are some sample unit tests:

C#:
 
Title and subtitle.
Anyway, I was afraid you would mix the two languages up.
 
5:09 PM
No point in giving both colons
##C#
##VB
 
Doesn't bug me.
If you want to edit, go ahead.
 
5:20 PM
1
Q: Remove Empty Argument Lists from Attributes

Hosch250I converted a C# analyzer for removing empty argument lists from attributes to be a C# and VB.NET analyzer: public override void Initialize(AnalysisContext context) { context.RegisterSyntaxNodeAction(AnalyzeSymbol, Microsoft.CodeAnalysis.CSharp.SyntaxKind.Attribute); } private void AnalyzeS...

 
@200_success let's get him hopelessly hooked :)
 
@janos You mean get him a Nice Answer badge?
And 100 rep?
 
not just any zombie.... damn, that's a monster zombie boss
 
We are almost there.
We just have to be sure to stop at 200 so he doesn't lose rep.
 
not sure if that will get him "hopelessly hooked", but a good first step I guess ;)
 
5:25 PM
Excel-cells vs. Arrays/memory There is a difference of 8s (wow), but for 100K records; this gap is kept constant as far as it becomes 40s for 500K. This is undoubtedly relevant, but I understand that this is not a surprise to anyone: memory is the fastest alternative in any programming language (and dealing with Excel cells is almost like dealing with local files). On the other hand.. 100K records in Excel is not precisely a too common situation. — varocarbas 53 secs ago
 
Congratulations on writing a very good answer on a large, old question! Feel free to join us in chat if you ever feel like it :) — Mast 13 secs ago
 
@Duga hey we're becoming a resource on SO!
7
 
@Hosch250 He got a shiny Necromancer out of it.
2 more till Nice Answer
TTG grab a movie
 
5:54 PM
@Hosch250 I don't understand at all why you can't do what I was recommending...
And your question doesn't clearly demark what is example and what is to be reviewed (and you should know better)
The tag on that question is probably a pretty terrible idea...
 
It is all to be reviewed, if you see something.
And there is vb.net in the unit test.
 
Yes and no.
 
Maybe I don't fully understand what you were recommending.
You say take one method, and pass it into two methods.
 
You are testing your C#'s ability to do stuff to VB.NET code you pass in.
But the VB.NET code isn't up for review
@Hosch250 No, I say write one method, and provide as many wrappers for it as you need to match the signatures you need to match.
 
OK.
So, wouldn't that end up duplicating all the code over again?
 
5:58 PM
No.
All of your code is in a single method. And you use if or switch or whatever means to handle the few divergent places, as you would normally do
 
Oh, OK.
 
All of your code is in the single method that the other "signature required methods" just wrap.
The methods that you must implement are 1-line calls to the main method that handles all of the logic.
 
Hi there! :)
 
And that method takes an extra argument
such as an enum
 
@MichaelGrünewald Hello! ;-)
 

« first day (855 days earlier)      last day (3177 days later) »