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!

«‹›1503 »

Soricidus: Oct 21, 2010; freedom-hating statist shill

Pavlov posted:

Even with a US keyboard, I would use a language that uses 「Japanese quotation marks」, just because I think they look spiffy.

e: Holy poo poo they got 『』, 《》, and【】 too. Think of all the special nesting types we could have.

sempython

# ? May 31, 2016 22:15

Adbot: ADBOT LOVES YOU

# ? May 17, 2024 20:41

Internet Janitor: May 17, 2008; "That isn't the appropriate trash receptacle."

Most of the time when I look at my old code unhappiness stems from realizing my old assumptions in designing the codebase were flawed, incomplete or no longer valid. Sometimes you need a first stab at a problem and some time to see these kinds of design mistakes.

For things that are naturally self-contained I'm more likely to look back and see a mixture of opportunities to improve and pleasantly clever bits.

# ? May 31, 2016 22:16

Soricidus: Oct 21, 2010; freedom-hating statist shill

Xerophyte posted:

Oh yes, the layouts themselves are horrors in a lot of ways. I maintain that the person who looked at the $ character and thought "that'd make a good special character in my spiffy programming language" was a bit blinkered, though.

It's not used in any good language I can think of, unless you count regex strings or non-user-facing stuff like name mangling.

# ? May 31, 2016 22:17

TooMuchAbstraction: Oct 14, 2012; I spent four years making
Waves of Steel
Hell yes I'm going to turn my avatar into an ad for it.; Fun Shoe

Soricidus posted:

It's not used in any good language I can think of, unless you count regex strings or non-user-facing stuff like name mangling.

I was going to counteroffer Perl and PHP, but you had to go and add that "good" modifier there.

(Disclaimer: I like Perl, but I would not want to work in a Perl shop)

# ? May 31, 2016 22:35

raminasi: Jan 25, 2005; a last drink with no ice

Soricidus posted:

It's not used in any good language I can think of, unless you count regex strings or non-user-facing stuff like name mangling.

String interpolation in C# 6.

# ? May 31, 2016 22:41

fritz: Jul 26, 2003

Soricidus posted:

You could try trigraphs.

Mega-rude.

# ? May 31, 2016 22:41

rjmccall: Sep 7, 2007; no worries friend; Fun Shoe

raminasi posted:

String interpolation in C# 6.

Also anonymous closure arguments in Swift. They're really nice when the closure is tiny.

# ? May 31, 2016 23:04

b0lt: Apr 29, 2005

M31 posted:

Or C!

code:

main(int argc, char *argv<::>) <%
    printf("Hello World");
%>

Trigraphs are going to be removed in C++17, with IBM being dragged kicking and screaming about their "totally real but we can't talk about them" EBCDIC-using customers. Hopefully C will follow?

# ? Jun 1, 2016 00:49

sarehu: Apr 20, 2007; (call/cc call/cc)

https://blog.devmastery.com/how-to-win-the-coding-interview-71ae7102d685

quote:

A palindrome is a word, phrase, number, or other sequence of characters which reads the same backward or forward. Allowances may be made for adjustments to capital letters, punctuation, and word dividers. Examples in English include �A man, a plan, a canal, Panama!�, �Amor, Roma�, �race car�, �stack cats�, �step on no pets�, �taco cat�, �put it up�, �Was it a car or a cat I saw?� and �No �x� in Nixon�.

Write the most efficient function you can that determines whether a given string is a palindrome.

Okay...

quote:

On this particular challenge, I am expecting many will use RegEx as a part of the solution.

Edit:

quote:

Before I start though, I have to say, if a company is going to hire a developer based solely and entirely on a piece of code the developer wrote in an interview, you probably don�t want to work there.

Ah, but of course we cannot forget culture fit.

sarehu fucked around with this message at 01:52 on Jun 1, 2016

# ? Jun 1, 2016 01:48

Plorkyeran: Mar 22, 2007; To Escape The Shackles Of The Old Forums, We Must Reject The Tribal Negativity He Endorsed

Unsurprisingly his example of a good solution is kinda poo poo.

# ? Jun 1, 2016 01:54

Pollyzoid: Nov 2, 2010; GRUUAGH you say?

b0lt posted:

Trigraphs are going to be removed in C++17, with IBM being dragged kicking and screaming about their "totally real but we can't talk about them" EBCDIC-using customers. Hopefully C will follow?

Those are digraphs though, and they're not going to be removed for a while.

# ? Jun 1, 2016 02:06

Absurd Alhazred: Mar 27, 2010; by Athanatos

Xerophyte posted:

On that note, it remains slightly annoying to me how English and US-centric a lot of the character conventions in programming are. ${}[]| and co aren't particularly easy to type on a good half of the world's keyboards, yet languages keep using them all over the drat place. Backticks are by far the worst in the swedish/nordic/german layouts, since ` is a chord resulting in a dead key meaning half the time you end up writing � instead of `a.

You might reply that I should use a US layout. I do, which has the cost that if I want to write something in my native language then that's a bunch of additional effort instead. The entire situation could be avoided by sticking to basic text-oriented symbols available everywhere but some jackass just had to give distinct meanings to all their special keys. I nurture a small, vindictive hope to some day introduce the � and � operators in a major language to ensure that person will be as mildly annoyed as I have been.

My other language is RTL and an entirely different alphabet. I have to use "alt-shift" to change languages multiple times a minute, when I'm talking to friends while looking up programming-related stuff. I manage, and so can you.

If you want things to be different, go back in time and make it so that the Swedes had invented and popularized programming or whatever. See if you can fend off the Israelis, Iranians, and Chinese who are trying to mess around with the same timeline, though. :shrug:

# ? Jun 1, 2016 02:38

Internet Janitor: May 17, 2008; "That isn't the appropriate trash receptacle."

Plorkyeran posted:

Unsurprisingly his example of a good solution is kinda poo poo.

Is this where we all trip over each other to make a better implementation and do point-by-point teardowns of solutions we don't like?

JavaScript code:

function isPalindrome(s) {
	var f = s.toLowerCase().replace(/[^a-z\d]/g, "");
	return f == f.split("").reverse().join("");
}

I think his insistence on handling bogus types with an explicit exception throw is silly; a non-string in this implementation will immediately explode for lack of a toLowerCase(). I also don't see any reason to complain about strings which are vacuously palindromes. Regular expressions are unnecessary for this type of problem, but I think they're reasonably defensible in light of the requirement to ignore characters beyond alphanumerics. I was surprised to discover that evidently JS regexes don't have a symbol equivalent to a-zA-Z0-9. This would also be simpler if JS Strings were genuinely Arrays.

# ? Jun 1, 2016 06:21

senrath: Nov 4, 2009; Look Professor, a destruct switch!

Internet Janitor posted:

I was surprised to discover that evidently JS regexes don't have a symbol equivalent to a-zA-Z0-9. This would also be simpler if JS Strings were genuinely Arrays.

Isn't that \w?

Edit: Double checking apparently that also includes underscores. Huh.

# ? Jun 1, 2016 06:42

FamDav: Mar 29, 2008

They didn't even parallelize the palindrome check

# ? Jun 1, 2016 06:45

Suspicious Dish: Sep 24, 2011; 2020 is the year of linux on the desktop, bro; Fun Shoe

I have no idea why, but I have consistently seen multiple cases where people where people try to check if something is a palindrome by doing:

JavaScript code:

var leftHalf = string.slice(0, string.length / 2);
var rightHalf = string.slice(string.length / 2);
return leftHalf == rightHalf.reverse();

I really don't know what it is.

# ? Jun 1, 2016 07:06

brap: Aug 23, 2004; Grimey Drawer

Writing a function that logs to the console when doing something very simple with no dependencies is just retarded.
That guy is definitely high on the sense that he's a PROFESSIONAL JavaScript developer and from sniffing his own farts.

# ? Jun 1, 2016 07:13

Jabor: Jul 16, 2010; #1 Loser at SpaceChem

Suspicious Dish posted:

I have no idea why, but I have consistently seen multiple cases where people where people try to check if something is a palindrome by doing:
JavaScript code:
var leftHalf = string.slice(0, string.length / 2);
var rightHalf = string.slice(string.length / 2);
return leftHalf == rightHalf.reverse();
I really don't know what it is.

Because they can see that the "obvious" solution of checking whether string == reverse(string) ends up checking each letter twice, and they're too clever for their own good.

# ? Jun 1, 2016 07:19

canis minor: May 4, 2011

quote:

�Do I need to handle unicode characters?�

Obviously the answer is no

isPalindrome("aaą");
> true

canis minor fucked around with this message at 07:39 on Jun 1, 2016

# ? Jun 1, 2016 07:36

Soricidus: Oct 21, 2010; freedom-hating statist shill

Absurd Alhazred posted:

My other language is RTL

oh god trigger warning that poo poo dude

I literally still have nightmares about bidi text handling, over a year after I last worked on anything that involves it

is the thing where numbers are always LTR actually good from your perspective? because it's the reason there is never a nice simple case

# ? Jun 1, 2016 07:52

Hammerite: Mar 9, 2007; And you don't remember what I said here, either, but it was pompous and stupid.; Jade Ear Joe

sarehu posted:

https://blog.devmastery.com/how-to-win-the-coding-interview-71ae7102d685

Okay...

I was going to say that he says that about regex, but then doesn't actually use regex as part of his model solution. But technically, he does - he uses it to filter out non-letter characters before he gets to the actual work of determining if it's a palindrome. I guess that's all he meant by "use regex as part of the solution". It's poor communication on his part, because he's unnecessarily drawing attention to a relatively unimportant preparatory step, which misleads the reader into thinking that the use of regex will be the meat of the solution. All round a fairly dubious article.

# ? Jun 1, 2016 10:15

Jewel: May 2, 2009

I don't know if this is the right one because I don't really want to think about it too hard, but I have seen people check for palindromes with regex recursion. This is a perl regex (only works in certain versions or something?)

code:

/^((.)(?1)\2|.?)$/

# ? Jun 1, 2016 10:22

TheresaJayne: Jul 1, 2011

Heres a new one,

A project decided to use bitmasks to store permissions in an int64.

They now have 65 permissions needed....

# ? Jun 1, 2016 10:56

feedmegin: Jul 30, 2008

TheresaJayne posted:

Heres a new one,

A project decided to use bitmasks to store permissions in an int64.

They now have 65 permissions needed....

Hope their compiler supports uint128_t!

# ? Jun 1, 2016 11:31

TheresaJayne: Jul 1, 2011

feedmegin posted:

Hope their compiler supports uint128_t!

VB.net

# ? Jun 1, 2016 11:40

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

Aren't palindromes like the classic example of the recursive version being simpler to read and easier to conceive. How is any regex, at all, better and more comprehensible than:

code:

bool isPal(int leftIndex, int rightIndex, String string){
    if leftIndex==rightIndex return true;
    if string.charAt(leftIndex)!=string.charAt(rightIndex) return false;
    return isPal(leftIndex++, rightIndex++, string);
}

void main{
   String str="taco cat";
   System.out.println((Bool)isPal(0,str.Length(),str).toString());
}

I'm sure there's some fucky syntax in there, but it's been ages since I got to use a real language.

The iterative version isn't that much harder to implement (even using that while(var++ && var--) syntax argued about earlier), and either is way better than going straight to regular expressions imo.

edit:

code:

ispalindrome(string) n begin,end,fail
	s string=$$stripspaces(string)
	s end=$l(string),begin=1
	f  d  q:fail  q:end=begin
	. i $e(string,begin)'=$e(string,end) s fail=1 q
	. s end=end-1,begin=begin+1
	q 'fail

The MUMPSorceress fucked around with this message at 16:14 on Jun 1, 2016

# ? Jun 1, 2016 16:06

TooMuchAbstraction: Oct 14, 2012; I spent four years making
Waves of Steel
Hell yes I'm going to turn my avatar into an ad for it.; Fun Shoe

The regex was used because part of the problem statement was to ignore non-letters (and casing). I admit I have trouble thinking of a better way than a regex to strip out undesirable characters, unless the language being used specifically has a function that happens to meet my needs.

# ? Jun 1, 2016 16:15

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

TooMuchAbstraction posted:

The regex was used because part of the problem statement was to ignore non-letters (and casing). I admit I have trouble thinking of a better way than a regex to strip out undesirable characters, unless the language being used specifically has a function that happens to meet my needs.

i can't think of any languages that don't have functions for moving to all caps, no caps, and stripping whitespace. stripping punctuation is a wrinkle i hadn't thought of for sure.

# ? Jun 1, 2016 16:23

rjmccall: Sep 7, 2007; no worries friend; Fun Shoe

LeftistMuslimObama posted:

Aren't palindromes like the classic example of the recursive version being simpler to read and easier to conceive. How is any regex, at all, better and more comprehensible than:
code:
bool isPal(int leftIndex, int rightIndex, String string){
    if leftIndex==rightIndex return true;
    if string.charAt(leftIndex)!=string.charAt(rightIndex) return false;
    return isPal(leftIndex++, rightIndex++, string);
}

Well, for one, it would terminate.

# ? Jun 1, 2016 16:26

New Yorp New Yorp: Jul 18, 2003; Only in Kenya.; Pillbug

LeftistMuslimObama posted:

i can't think of any languages that don't have functions for moving to all caps, no caps, and stripping whitespace. stripping punctuation is a wrinkle i hadn't thought of for sure.

If you also take into account unicode, it suddenly becomes a big pain in the rear end, too. Like combining characters:

Les Mis\u030erables = Les Mis�rables

Reverse it!

selbare\u030siM seL = selbareśiM seL

Oops, the accent is over the wrong character now.

(Example stolen from Jon Skeet: https://codeblog.jonskeet.uk/2009/11/02/omg-ponies-aka-humanity-epic-fail/)

# ? Jun 1, 2016 16:29

OddObserver: Apr 3, 2009

TooMuchAbstraction posted:

The regex was used because part of the problem statement was to ignore non-letters (and casing). I admit I have trouble thinking of a better way than a regex to strip out undesirable characters, unless the language being used specifically has a function that happens to meet my needs.

An if statement? Does everything have to be a precanned library function?

# ? Jun 1, 2016 16:31

The MUMPSorceress: Jan 6, 2012; ^SHTPSTS; Gary’s Answer

rjmccall posted:

Well, for one, it would terminate.

Lol, I always gently caress that up.

# ? Jun 1, 2016 16:32

TooMuchAbstraction: Oct 14, 2012; I spent four years making
Waves of Steel
Hell yes I'm going to turn my avatar into an ad for it.; Fun Shoe

OddObserver posted:

An if statement? Does everything have to be a precanned library function?

No, I was saying that in the absence of some kind of stripNonLetters() library method, a regex would be the only other good option for removing non-letter characters from a string that I can think of. And in fact I wouldn't be surprised if such a library method were just a thin wrapper around a regex anyway.

I mean, sure, you can iterate over every character and test if the char value of the character falls between 65 and 90 or 97 and 122 (or look up if the character is in a whitelist by other methods), assuming you're happy assuming ASCII. I don't think that's a superior solution though.

# ? Jun 1, 2016 16:36

qntm: Jun 17, 2009

Unicode has no notion of string reversal, so I think it's fair to treat the input as an opaque sequence of code points or whatever.

# ? Jun 1, 2016 16:54

OddObserver: Apr 3, 2009

Who said you need to mutate the string at all? (Well, C-form normalization probably is a good idea, yes).

Edit: what I mean is, whole thing is easily doable in a manner similar to the "collapsing walls" quick sort partition function, just w/o swaps.

OddObserver fucked around with this message at 17:05 on Jun 1, 2016

# ? Jun 1, 2016 16:58

eth0.n: Jun 1, 2012

Thought I'd try doing it the "proper" C++ way, using iterators, and avoid string copying and redundant comparisons. Reverse iterators turn out to be kind of a pain. Would have been nice if they could directly be compared to forward iterators, or if there were a consistent way to "toggle" the reverse-ness of an iterator.

code:

template<typename Iter>
Iter skip_nonletter(Iter i, Iter limit)
{
  while(i <= limit)
  {
    char c = tolower(*i);
    if(c >= 'a' && c <= 'z')
      break;
    ++i;
  }
  return i;
}

bool palindrome(const std::string &s)
{
  auto left = s.cbegin();
  auto right = s.crbegin();
  for(;;)
  {
    left = skip_nonletter(left, right.base());
    right = skip_nonletter(right, std::reverse_iterator<decltype(left)>(left));
    if(left > right.base())
      break;
    if(tolower(*left) != tolower(*right))
      return false;
    ++left;
    ++right;
  }
  return true;
}

This isn't something I'd be able to cough up in an interview, though. Probably would just go C-style, with indexes, and somewhat sloppy near-duplication of the "skip nonletter" code (i.e., same, but reversed direction).

eth0.n fucked around with this message at 17:15 on Jun 1, 2016

# ? Jun 1, 2016 17:13

xzzy: Mar 5, 2009

I can count the number of times I've had to reverse a string in the past 20 years on zero fingers.

Teachers should probably do a better job coming up with assignments that actually have application in the real world. Want to teach regex? Make them parse logs for specific data. Want to teach recursion? Implement quicksort or something.

# ? Jun 1, 2016 17:24

leper khan: Dec 28, 2010; Honest to god thinks Half Life 2 is a bad game. But at least he likes Monster Hunter.

Using a regex is still bad because you need to at minimum perform a bunch of writes and likely an allocation. You have a fixed number of valid characters to test against, so you can just skip everything that isn't valid in constant time per character and run through with a forward and backward iterator in O(N) where N is the length of the input. Testing against both the character's upper/lower also takes constant time per test. As you need to at minimum read the whole string to check the characters, you're already at the best asymptotic time with reasonable use of memory and you can get on with your day and doing something relevant.

# ? Jun 1, 2016 17:50

TooMuchAbstraction: Oct 14, 2012; I spent four years making
Waves of Steel
Hell yes I'm going to turn my avatar into an ad for it.; Fun Shoe

leper khan posted:

Using a regex is still bad because you need to at minimum perform a bunch of writes and likely an allocation. You have a fixed number of valid characters to test against, so you can just skip everything that isn't valid in constant time per character and run through with a forward and backward iterator in O(N) where N is the length of the input. Testing against both the character's upper/lower also takes constant time per test. As you need to at minimum read the whole string to check the characters, you're already at the best asymptotic time with reasonable use of memory and you can get on with your day and doing something relevant.

I can write a regex to strip invalid characters a hell of a lot faster than I can write the relevant whitelist and loop to do it manually. Is the manual approach more efficient? Probably. But I'll be getting on with my day doing something relevant faster if I make full use of the available tools.

# ? Jun 1, 2016 18:14

Adbot: ADBOT LOVES YOU

# ? May 17, 2024 20:41

Absurd Alhazred: Mar 27, 2010; by Athanatos

Soricidus posted:

oh god trigger warning that poo poo dude

I literally still have nightmares about bidi text handling, over a year after I last worked on anything that involves it

is the thing where numbers are always LTR actually good from your perspective? because it's the reason there is never a nice simple case

In Hebrew numbers are spoken LTR/big-endian, so it works out well for typing. I have just realized, though, that people using physical typewriters must have been writing numbers RTL/little-endian.That must have been an irritation, both getting used to it, and then getting used to the new way.

# ? Jun 1, 2016 18:25

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!

«‹›1503 »