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 »

Presto: Nov 22, 2002; Keep calm and Harry on.

Hey, at least whoever wrote the number-to-words function did a good job though. It correctly handled a number in the trillions.

# ? Feb 13, 2022 21:10

Adbot: ADBOT LOVES YOU

# ? May 27, 2024 00:25

Foxfire_: Nov 8, 2010

I'm from a country that uses whichever trillion scale you don't use and say the number-to-words function is garbage! :argh:

# ? Feb 14, 2022 00:10

Bonfire Lit: Jul 9, 2008; If you're one of the sinners who caused this please unfriend me now.

TypeScript code:

function checkValue(input: string): number | boolean {
  if (someRegexp.test(input)) {
    return NaN;
  }
  return true;
}

# ? Feb 21, 2022 10:24

Votlook: Aug 20, 2005

Bonfire Lit posted:

TypeScript code:

function checkValue(input: string): number | boolean {
  if (someRegexp.test(input)) {
    return NaN;
  }
  return true;
}

This malice, not stupidity,

# ? Feb 21, 2022 13:52

D34THROW: Jan 29, 2012; RETAIL RETAIL LISTEN TO ME BITCH ABOUT RETAIL

Bonfire Lit posted:

TypeScript code:

function checkValue(input: string): number | boolean {
  if (someRegexp.test(input)) {
    return NaN;
  }
  return true;
}

I dont know TypeScript what the gently caress is happening here

# ? Feb 21, 2022 13:53

Tei: Feb 19, 2011

D34THROW posted:

I dont know TypeScript what the gently caress is happening here

I don't know either.

Seems to be some sort of validation function that can return NaN if a regex checks, otherwise return true.

I guess the reason why is posted here is having a function that can return only two values true and NaN, and not using a boolean for that.

It would made more sense to me to return null instead of NaN, or simply false.

null being defined in database realms has "I dunno, man".

# ? Feb 21, 2022 14:04

enki42: Jun 11, 2001; #ATMLIVESMATTER

Put this Nazi-lover on ignore immediately!

It's just about the worst named function I've ever seen. What it's supposed to do depends heavily on someRegExp which I'm assuming is a constant. If it was something that checked whether something is a number I suppose it could be something to test whether a string can be parsed to a number, although who the hell knows why you'd return NaN if it doesn't.

# ? Feb 21, 2022 14:19

Phobeste: Apr 9, 2006; never, like, count out Touchdown Tom, man

enki42 posted:

It's just about the worst named function I've ever seen. What it's supposed to do depends heavily on someRegExp which I'm assuming is a constant. If it was something that checked whether something is a number I suppose it could be something to test whether a string can be parsed to a number, although who the hell knows why you'd return NaN if it doesn't.

Because then it's not a number op couldn't be simpler

# ? Feb 21, 2022 14:22

enki42: Jun 11, 2001; #ATMLIVESMATTER

Put this Nazi-lover on ignore immediately!

I don't know how you write the signature for that return value in Typescript and not realize that you're doing something very, very wrong.

# ? Feb 21, 2022 14:25

DoctorTristan: Mar 11, 2006; I would look up into your lifeless eyes and wave, like this. Can you and your associates arrange that for me, Mr. Morden?

It�s the sort of thing I�d write if I were quitting tomorrow and hated my coworkers.

# ? Feb 21, 2022 14:28

Jazerus: May 24, 2011

D34THROW posted:

I dont know TypeScript what the gently caress is happening here

it's bad enough that the language hands you NaNs occasionally, writing a function that does it intentionally is a war crime

# ? Feb 21, 2022 15:27

ExcessBLarg!: Sep 1, 2001

Bonfire Lit posted:

TypeScript code:

function checkValue(input: string): number | boolean {
  if (someRegexp.test(input)) {
    return NaN;
  }
  return true;
}

I'm not familiar with TypeScript either so the only way this makes sense to me is if these assumptions are true and it's being used in this context:

1. NaN coerces to false in boolean context.
2. This function is being called as a short circuit OR operation.
3. The input is a string representation in of a rational (fraction) that needs to be converted (evaled) into a floating-point value.
4. The purpose is to catch an integer divide by zero before evaling the string, to avoid a divide by zero exception and instead provide a reasonable (if imprecise) floating-point response.

# ? Feb 21, 2022 18:01

Mata: Dec 23, 2003

I imagine if you are writing that sort of function you're probably also playing fast and loose with JS == operator wherever that returned NaN shows up

# ? Feb 21, 2022 18:13

Jazerus: May 24, 2011

ExcessBLarg! posted:

I'm not familiar with TypeScript either so the only way this makes sense to me is if these assumptions are true and it's being used in this context:

1. NaN coerces to false in boolean context.
2. This function is being called as a short circuit OR operation.
3. The input is a string representation in of a rational (fraction) that needs to be converted (evaled) into a floating-point value.
4. The purpose is to catch an integer divide by zero before evaling the string, to avoid a divide by zero exception and instead provide a reasonable (if imprecise) floating-point response.

NaN does coerce to false, but JS/TS don't freak out when you divide by zero, they return infinity, a special number that is sort of but not exactly like mathematical infinity, which is a reasonable if imprecise response to dividing by zero. feeding NaN into something that expects a number, however, will cause an error.

e: which, idk, maybe you'd rather get an error than send infinity out into the world in this case, hard to say

Jazerus fucked around with this message at 18:23 on Feb 21, 2022

# ? Feb 21, 2022 18:20

Ola: Jul 19, 2004

I'm not familiar with Typescript either, but I'm pretty sure the type signature means it's a union type so it returns either a number or a boolean. It's fine in that regard, apart from being horrible in all regards. I'm guessing it's written to check if it's a number and the caller does a check on != NaN, for some strange reason that probably involves lots of broth and many cooks.

# ? Feb 21, 2022 19:19

DoctorTristan: Mar 11, 2006; I would look up into your lifeless eyes and wave, like this. Can you and your associates arrange that for me, Mr. Morden?

Ola posted:

�check if it's a number and the caller does a check on != NaN, �

Hahaha I bet it does exactly this lol

# ? Feb 21, 2022 19:41

pokeyman: Nov 26, 2006; That elephant ate my entire platoon.

Today I learned NaN is not a valid literal type in TypeScript, further confirming the excellent taste of its maintainers.

# ? Feb 21, 2022 20:09

canis minor: May 4, 2011

NaN is in fact a Number

# ? Feb 21, 2022 20:47

Bonfire Lit: Jul 9, 2008; If you're one of the sinners who caused this please unfriend me now.

Jazerus posted:

it's bad enough that the language hands you NaNs occasionally, writing a function that does it intentionally is a war crime

You'll be glad to hear then that the return value of that function is never used except in boolean context (i.e., if (checkValue(foo)) { do something })

# ? Feb 21, 2022 20:57

Ola: Jul 19, 2004

Bonfire Lit posted:

You'll be glad to hear then that the return value of that function is never used except in boolean context (i.e., if (checkValue(foo)) { do something })

So kind of what I guess, only the other way around. Why, for the love of all that is holy, not if (someRegexp.test(input)) { do something } ?

# ? Feb 21, 2022 21:21

Bonfire Lit: Jul 9, 2008; If you're one of the sinners who caused this please unfriend me now.

Beats me! I've elected not to check source control because I suspect if I do I'll just get salty at whoever wrote that and whoever reviewed it afterwards.

# ? Feb 21, 2022 21:29

Jazerus: May 24, 2011

Bonfire Lit posted:

You'll be glad to hear then that the return value of that function is never used except in boolean context

for now

# ? Feb 21, 2022 21:35

Foxfire_: Nov 8, 2010

ExcessBLarg! posted:

4. The purpose is to catch an integer divide by zero before evaling the string, to avoid a divide by zero exception and instead provide a reasonable (if imprecise) floating-point response.

Surprise! Javascript/typescript doesn't have integers

# ? Feb 21, 2022 22:00

DoctorTristan: Mar 11, 2006; I would look up into your lifeless eyes and wave, like this. Can you and your associates arrange that for me, Mr. Morden?

Bonfire Lit posted:

Beats me! I've elected not to check source control because I suspect if I do I'll just get salty at whoever wrote that and whoever reviewed it afterwards.

So this person is also in the habit of eval()-ing strings, you say?

# ? Feb 21, 2022 22:08

BigPaddy: Jun 30, 2008; That night we performed the rite and opened the gate.
Halfway through, I went to fix us both a coke float.
By the time I got back, he'd gone insane.
Plus, he'd left the gate open and there was evil everywhere.

Always fun when you solve a problem on AWS by making a copy of the function that is not working, set it up the same, the code is the same, the permissions are the same and it works when the previous one you copied it from and should be identical throws errors with no message.

# ? Feb 21, 2022 23:10

NihilCredo: Jun 6, 2011; iram omni possibili modo preme:
plus una illa te diffamabit, quam multæ virtutes commendabunt

Foxfire_ posted:

Surprise! Javascript/typescript doesn't have integers

Says you.

# ? Feb 22, 2022 00:56

pokeyman: Nov 26, 2006; That elephant ate my entire platoon.

Bonfire Lit posted:

Beats me! I've elected not to check source control because I suspect if I do I'll just get salty at whoever wrote that and whoever reviewed it afterwards.

You can just say it was you

# ? Feb 22, 2022 01:59

Tei: Feb 19, 2011

I did this on a javascript console.

NaN?true:false;

it returns false, so NaN evaluates to false, so thats why that function works. while NaN is not a boolean, evaluated in a boolean context it works like a false.

this is C spirit working from inside javascript, convenient if tricky

# ? Feb 22, 2022 10:22

Sagacity: May 2, 2003; Hopefully my epitaph will be funnier than my custom title.

Tei posted:

this is C spirit working from inside javascript, convenient if tricky

more like a spirit like Jack Daniels is

# ? Feb 22, 2022 10:25

Ola: Jul 19, 2004

Tei posted:

I did this on a javascript console.

NaN?true:false;

it returns false, so NaN evaluates to false, so thats why that function works. while NaN is not a boolean, evaluated in a boolean context it works like a false.

this is C spirit working from inside javascript, convenient if tricky

It isn't in a boolean context though, it's in a union type context where it can be either a number or a boolean. Union types are cool and good in functional languages, but Typescript isn't one of those.

NaN is extra super falsy in Javascript so you really wouldn't want to use it for anything.

code:

false==false
true
NaN==NaN
false
NaN===NaN
false

# ? Feb 22, 2022 10:39

Xarn: Jun 26, 2015

That's just NaN behaviour in ieee 754.

# ? Feb 22, 2022 11:00

D34THROW: Jan 29, 2012; RETAIL RETAIL LISTEN TO ME BITCH ABOUT RETAIL

How the hell is a NaN not strictly equal to NaN :psyduck:

# ? Feb 22, 2022 14:02

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

D34THROW posted:

How the hell is a NaN not strictly equal to NaN

How could it be? It holds a meaning outside its bit set.

# ? Feb 22, 2022 14:11

Tei: Feb 19, 2011

Ola posted:

code:

false==false
true
NaN==NaN
false
NaN===NaN
false

:-O

what the gently caress. I mean... null in databases is probably similar, but what the gently caress

# ? Feb 22, 2022 14:14

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

Tei posted:

:-O

what the gently caress. I mean... null in databases is probably similar, but what the gently caress

This is standard behavior for floats. Anything else would be against the spec and extremely wrong. What were you expecting?

# ? Feb 22, 2022 14:18

ToxicFrog: Apr 26, 2008

D34THROW posted:

How the hell is a NaN not strictly equal to NaN

First time, eh?

# ? Feb 22, 2022 14:22

ynohtna: Feb 16, 2007; backwoods compatible; Illegal Hen

Tei posted:

:-O

what the gently caress. I mean... null in databases is probably similar, but what the gently caress

It's NaN-defined behaviour.

# ? Feb 22, 2022 14:22

nielsm: Jun 1, 2009

NaN by definition represents an unknowable value, so you have to assume that two NaN values are different.
Usually SQL NULL represents the absence of a value, while NaN represents that a value is present but unknowable.

# ? Feb 22, 2022 14:32

Tei: Feb 19, 2011

leper khan posted:

This is standard behavior for floats. Anything else would be against the spec and extremely wrong. What were you expecting?

https://www.youtube.com/watch?v=6ZpaWOLjWx0

For it to explode comically, like in a old movie, when computers where still called "electronic brains".

# ? Feb 22, 2022 14:39

Adbot: ADBOT LOVES YOU

# ? May 27, 2024 00:25

OddObserver: Apr 3, 2009

Signaling NaNs are indeed also a thing, but not in JS.

# ? Feb 22, 2022 14:42

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 »