Coding Horrors: You can gather all your technical debt into one easy framework!

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Cavern of COBOL > Coding Horrors: You can gather all your technical debt into one easy framework!

«‹›2 »

Victor: Jun 18, 2004

tef posted:

See, where you might use "result" I would use bln_flg_tst_cnd_args_x.

You almost made me spit coffee over my new, broken-in MS 4000 keyboard!

# ¿ Mar 22, 2008 06:16

Adbot: ADBOT LOVES YOU

# ¿ Apr 28, 2024 19:11

Victor: Jun 18, 2004

code:

string clientStr = this
    .publishToClients
    .OfType<PublishToSetting>()
    .Select(c => string.Format("{0}:{1}:{2}:{3}", c.Alias, c.Ip, c.Port, c.TheProtocol))
    .Join(",");

(string Join(string delimiter) is a method I made, because MS didn't include it by default, for some odd reason.)

# ¿ Mar 26, 2008 18:47

Victor: Jun 18, 2004

_aaron posted:

Hahaha, that's pretty awesome. But correct me if I'm wrong here: most of that stuff is .Net 3.5 stuff, right? We are still using vs2005 (but will be upgrading soon), so that doesn't apply to our stuff. When we do make the switch, I will certainly keep this in mind.

Yup, .NET 3.5, C# 3.0.

# ¿ Mar 26, 2008 21:06

Victor: Jun 18, 2004

Notice how the signature between what I posted differs from what you posted. Moreover, mine works on IEnumerable<string> and is an extension method. Here's the definition:

code:

public static string Join(this IEnumerable<string> strings, string delimiter)
{
    return strings.Any()
        ? string.Join(delimiter, strings as string[] ?? strings.ToArray())
        : "";
}

# ¿ Mar 27, 2008 10:03

Victor: Jun 18, 2004

I'm all for a few intelligent conventions like what you've just posted (I'm not sure if the actual ones you posted are good or not). Consistency is really the important thing, within any given code base. And don't make it drastically different from other code bases you're going to be working on. Give the brain fewer differences so that the differences it does process are the relevant ones.

# ¿ Mar 29, 2008 08:06

Victor: Jun 18, 2004

nebby, what if the programmers with whom you work just suck at giving proper context? Also, you repeatedly forget to note that this is much more important for dynamic languages, as things are less clear there. Have you looked into coding for the military, using ADA? It gets fairly close to your approach, but has compiler-enforced type safety. You can only get speed if you divide distance by time (or integrate acceleration by time, etc.).

Maybe I missed it, but I do not remember you ever explaining why your desire for Hungarian whatever is in the distinct minority of programmers. Does your approach really allow you to code more quickly and more accurately than other people, on average? (Take into account learning curve here!) To what extent are the problems you encounter, unique to your problem domain, or at least a very, very small fraction of programming?

The more I think about this, the more you must address the above if you want to stop digging/get out of the hole you've dug. Right now, you're some weird freak in most of our minds. I'm cutting you more slack than most, because I have very little respect for experience and track record, if I cannot understand the why behind the experience and track record. However, most people do, and you, my friend, have demonstrated zero.

# ¿ Mar 30, 2008 18:58

Victor: Jun 18, 2004

I don't think we should stop at making programmers manage their own type systems in symbol names. I think they should have to code in assembly. Screw that: ones and zeros. Actually, what would happen if we made programmers code mechanically, maybe via some sort of punch card system that is about as reliable as the punch cards used in the Florida election? Surely, in that case, the bug count would go way down.

# ¿ Mar 31, 2008 16:22

Victor: Jun 18, 2004

rotor posted:

We're gonna need bigger keyboards.

Apparently it's about as hard to go Hungarian (^_{_System} Apps) as to learn a new keyboard layout, so this might not be a half bad idea!

# ¿ Mar 31, 2008 17:36

Victor: Jun 18, 2004

yaoi prophet posted:

Is there a das keyboard version?

gently caress that, I want an OLED version.

# ¿ Mar 31, 2008 18:30

Victor: Jun 18, 2004

Clearly, return is a function. :psyduck:

# ¿ Apr 2, 2008 00:41

Victor: Jun 18, 2004

dustgun posted:

I should mention, just for the sake of mentioning, that this was written by a coworker of a friend of mine. I changed the method and variable names, but, it's real live code somewhere.

Are you a member of the NRA yet?

# ¿ Apr 2, 2008 05:59

Victor: Jun 18, 2004

I dunno man, insert some underscores into the obfuscated version and it starts looking like Hungarian...

# ¿ Apr 2, 2008 06:24

Victor: Jun 18, 2004

Clearly the input should be s and the output should be ret. :colbert:

# ¿ Apr 2, 2008 16:34

Victor: Jun 18, 2004

I reserve r for Regex instances.

# ¿ Apr 2, 2008 18:33

Victor: Jun 18, 2004

TRex EaterofCars posted:

Boo!

I use re for regex. It's far less common than returning a value, so it can get the extra letter.

Hrm, you might have just started to convert me from ret to r. Nobody tell nebby!

That Turkey Story posted:

he said that he was using the parenthesis not because he thought it was a function but because he wasn't sure of the order of operations.

Cool, I approve of this. It sounds like this guy does something extremely rare -- it seems like he actually tries to be aware of what he doesn't know!

quote:

(though honestly who really cares, some extra parenthesis never hurt anyone).

I'm one of those guys who doesn't want extraneous information screwing with my precious brain cycles.

# ¿ Apr 3, 2008 23:18

Victor: Jun 18, 2004

I general, I'd keep away from stuff like s.IsNullOrEmpty(), but I think I would make an except for that specific case. Especially since the name itself acknowledges that null is a valid "argument".

# ¿ Apr 7, 2008 20:12

Victor: Jun 18, 2004

floWenoL posted:

What happens when s is null?

The earth implodes.

code:

catch (NullReferenceException)
{
    Planets.Earth.FormBlackHole();
}

# ¿ Apr 8, 2008 14:10

Victor: Jun 18, 2004

Gotta protect against FILE_NOT_FOUND!

# ¿ Apr 10, 2008 22:53

Victor: Jun 18, 2004

You mean, a programming forum.

# ¿ May 13, 2008 20:20

Victor: Jun 18, 2004

It's best to have boolean variable names that sound boolean, like hidden (or even isHidden). I prefix nonpublic class fields with _ -- it's simple and to the point. With respect to calling a method inside a class, I'm a bit conflicted: if there's no this., the method invoked has to be one belonging to the enclosing type, at least in C#.

# ¿ May 15, 2008 16:47

Victor: Jun 18, 2004

I think this looks best:

code:

if( true == foo.equals(foo2) == true)

(Just don't do the same with false.)

# ¿ May 16, 2008 00:31

Victor: Jun 18, 2004

We should just rename DISTINCT to FIX.

# ¿ Jun 2, 2008 19:33

Victor: Jun 18, 2004

dwazegek posted:

If the key doesn't exist the dictionary will throw an exception, it won't return null.

Note if it is a NameValueCollection:

NameValueCollection.Item Property posted:

This property returns a null reference (Nothing in Visual Basic) in the following cases: 1) if the specified key is not found; and 2) if the specified key is found and its associated value is a null reference (Nothing in Visual Basic). This property does not distinguish between the two cases.

I learned something new: NameValueCollections allow .Add to be called multiple times with the same key, in which case a comma-delimited list is returned upon accessing that key.

# ¿ Jun 25, 2008 13:47

Victor: Jun 18, 2004

Unless a throw was omitted from the catch{} statement above, then adding the catch{} changes the semantics to the exception being handled. Not good!

# ¿ Jul 8, 2008 22:29

Victor: Jun 18, 2004

Bonus posted:

code:
  catch (Exception e)
  {
    transaction.Rollback();
    throw e;
  }
then any exception will be caught but the type of the exception propagated upwards will stay in tact. At least I think so, my Java is pretty rusty.

The above is horrible, as you lose the stack trace. Either throw;, or throw new DeleteFailedException(e); so that the stack trace is maintained via InnerException.

# ¿ Jul 10, 2008 17:30

Victor: Jun 18, 2004

HappyHippo posted:

Hmm

code:

sometype GetInstance(string Reg)
{
      switch (Reg)
      {
            case "first_field":
                  return first_field;
            case "second_field":
                  return second_field;
            case "third_field":
                  return third_field;
            case "fourth_field":
                  return fourth_field;
            case "fifth_field":
                  return fifth_field;
             .
             .
             .
      }
}

For 11,000 lines. This is in C#. Ignoring problems with the program's structure, I was able to turn this into about 5 lines of code with reflection.

Wait, so you have a class with 5,500 fields?

# ¿ Sep 8, 2008 15:36

Victor: Jun 18, 2004

BigRedDot posted:

Seriously, what is attempting to iterate past the end of a sequence, if not an exceptional circumstance?

This is standard operating procedure. In many, many different APIs, Read() will return 0 or null or some other sentinel value that there are no more values to consume. C++'s STL has iterators that evaluate to .end() when iteration is done -- .end() is not actually a valid element, it is one past the end of the sequence. The very nature of loops in [most] programming languages dictates that being able to get a last element that will never need to be processed (instead, it provides a sentinel value for a boolean "break out of this loop" test) is very useful. A fake, end-of-sequence element is occasionally added to facilitate this. No, exceptions are an ugly solution here, but may be the most elegant as stated above. I'm not sure whether static analysis makes this easier/possible, but the C# compiler implements iterators as state machines. No exceptions are required.

Victor fucked around with this message at 22:03 on Nov 8, 2008

# ¿ Nov 8, 2008 20:28

Victor: Jun 18, 2004

The real WTF is in several online tutorials that teach you to use a StringBuilder for instances such as the above, instead of letting the compiler automatically combine all the string constants into one. It's pretty sad.

# ¿ Nov 19, 2008 20:12

Victor: Jun 18, 2004

The real WTF is that Java doesn't provide single, simple methods to read/write all text/lines, like .NET does. Hmmmm, maybe just wanting to get a file's contents without caring about how is a common enough operation to provide a shortcut in the standard library?

P.S. Another problem with calling readLine in two places is that continue will screw you over.

# ¿ Nov 21, 2008 02:57

Victor: Jun 18, 2004

tef posted:

What if b[] is being modified in a different thread?

[list=1][*]I'm pretty sure you get a new b -- I know .NET doesn't resize arrays in-place and I doubt Java does either.[*]Assuming in-place array modification, the code isn't threadsafe anyway.[/list]

# ¿ Nov 21, 2008 03:30

Victor: Jun 18, 2004

ashgromnies posted:

who the gently caress in the history of computing thought, "ah, yes, xml is a decent markup language for organizing data... let's make code out of it"

When compared to some of the other utter and absolute crap out there, like healthcare's 837 file format, XML would be a godsend. One site calls it "overwhelmingly difficult to understand", and yet it's just supposed to be a hierarchical data format. Yes, there is much better for specific cases, but there is also much worse, and at least nubs can use XML...

# ¿ Nov 28, 2008 03:11

Victor: Jun 18, 2004

Not to mention reentrancy issues...

# ¿ Dec 4, 2008 20:30

Victor: Jun 18, 2004

For those of you with SQL2005, NEWSEQUENTIALID() was created for you:

MSDN posted:

Creates a GUID that is greater than any GUID previously generated by this function on a specified computer.

# ¿ Dec 5, 2008 03:57

Victor: Jun 18, 2004

O(1) for very large values of 1.

# ¿ Dec 5, 2008 06:03

Victor: Jun 18, 2004

Ledneh posted:

Spent the last hour reading this thread and now I'm extraordinarily worried I'm going to end up in here some day. There were so many in here that I didn't understand without explanatory posts

Anyone in here who claims to have never written code that could be submitted to this thread is a pretender*. The key is that you continue to learn and figure out ways to be less retarded.

* or hiiiilarious

# ¿ Dec 6, 2008 03:51

Victor: Jun 18, 2004

Haha, there's more WTFage -- if the getter is called without a properly cased string, not only will the property fail to be set, but it won't complain, as it has no default: section. WTF indeed!

# ¿ Dec 16, 2008 18:25

Victor: Jun 18, 2004

The WTF here is probably writing a logging framework instead of using one -- or did you really need to write your own?

# ¿ Dec 16, 2008 18:40

Victor: Jun 18, 2004

twodot posted:

This is kind of bizarre in my opinion. With a sufficiently strong type system you can claim any information is part of the typing. There are trade offs involved with deciding what information you put into a typedef and what information you put into a variable name. Personally, I think it can be useful to put information like whether a variable counts characters or bytes into the variable name rather than as a comment in the variable declaration or as a named type. Jumping between usages and definitions has costs especially in more object oriented code where definitions are non-local.

Hungarian notation isn't anything more than a standardized method of encoding information about a variable into its name. This is something that is always done when a variable name is created, standardizing it isn't intrinsically bad. Though most people's usages of Hungarian notation is very bad.

I think this is the best way to frame the "pro-Hungarian" perspective that I've ever seen. I'd create a new thread, but I have little confidence that the discussion would stay at the level that has been maintained for a few posts here. I'm not saying that I agree with twodot, but one would be retarded not to acknowledge that the point brings up valid concerns.

# ¿ Dec 30, 2008 19:03

Victor: Jun 18, 2004

Ryouga Inverse posted:

I challenge you to prove to me that mysql_real_escape_string() takes care of absolutely every possible SQL injection attack, including ones that may be discovered later.

This discussion fails because PHP and MySQL are so hosed up that if you are not paranoid enough, that if you don't pay absolute attention to security, you will lose and be hax0red. If you do "abc...'...wut".Replace("'", "''") in .NET, you are guaranteed that will work well. No encoding poo poo (everything is UTF-16 in .NET strings), nothing. There is no way you can fail unless you don't do that .Replace religiously. Now, few people are rigorous enough to do this, so they should suck it up and use parameterized queries. Whether you use prepared queries is immaterial -- that's only one way to parameterize in a safe way. I cannot believe there is so much loving bullshit on this issue. I can only attribute it to PHP and MySQL being lovely pieces of software. If you aren't anal about doing stuff right in the core layer, the layer upon which millions of people build, then you are worthless. You deserve to languish in the hell of bad code and ugly frameworks. I'm not sure I would work on PHP and MySQL if you paid me $1 million a year. They suck that much. The fact that plenty of people use these is irrelevant and you know it (or you are a retarded loser who never leaves his bubble).

I was just visiting this insults page and teapot's quote really resonated with me:

teapot posted:

No, it merely poisons minds, making programmers incapable of writing high-quality software and developing new ideas. It cripples their model of thinking by pushing them toward the optimal way of programming for severely mis-designed Microsoft products. Destruction of human minds' ability to improve is something that I don't take lightly.

I used to doubt this, but now I could not agree more. Except that Microsoft does not hold the patent on poo poo generation.

Victor fucked around with this message at 07:58 on Jan 8, 2009

# ¿ Jan 8, 2009 07:28

Adbot: ADBOT LOVES YOU

# ¿ Apr 28, 2024 19:11

Victor: Jun 18, 2004

I guess it's just hard for old people to be truly offensive other than, you know, being old and crufty and smelly with bad breath and only knowing ancient languages that stunt brand new minds. Since that's the kind of person who probably wrote up that site, well, you're not special enough to make it in.

# ¿ Jan 8, 2009 08:36

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > The Cavern of COBOL > Coding Horrors: You can gather all your technical debt into one easy framework!

«‹›2 »