terrible programmers: my god, it's full of chars

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > terrible programmers: my god, it's full of chars

«‹›1446 »

hackbunny: Jul 22, 2007; I haven't been on SA for years but the person who gave me my previous av as a joke felt guilty for doing so and decided to get me a non-shitty av

hackbunny posted:

on the client side Xcode works great, except for a few wrinkles. excessively minimalist UI maybe but I've seen worse. yet another command language to learn and no straightforward documentation (Microsoft has the best documentation. every single debugger command, every single error code or BSOD code, every single compiler warning or optimization switch is documented)

just found this lol: http://fuckingclangwarnings.com/

paging rjmccall, why don't you guys write some straightforward documentation of all the -W, -f, -O, -m etc. flags? why does the --help text for -fsanitize tell me to look in the manual, instead of listing the available sanitizers? why does --version print the version of llvm but not the version of clang itself? why does -fsanitize=address not work on ios (builds fine, but can't find the sanitizer's dll at runtime), but if I enable the address sanitizer from the workspace scheme it does? why isn't -fstack-protector on by default in Xcode, but -D_FORTIFY_SOURCE is? how bad is -fstack-protector-all really?

# ? Feb 19, 2016 22:35

Adbot: ADBOT LOVES YOU

# ? May 11, 2024 16:21

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

today i am the terrible programmer. i have a branch (we use svn), and i cannot get one of my files to commit to trunk. ive made whitespace changes, done a force merge, nothing gets my changes into the trunk. and its all made worse by the fact that we actually have like 3 trunks i need to move this through and they're all in different states. urgh.

# ? Feb 19, 2016 23:28

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

hackbunny posted:

paging rjmccall, why don't you guys write some straightforward documentation

documentation is probably the biggest weak point of anything llvm-related. clang's documentation of their language extensions is great, but the documentation of the command-line options is nearly nonexistent. libclang's api is nice, but the docs are basically useless due to being incredibly incomplete and frequently wrong or misleading. lldb basically just doesn't have documentation (there's an introduction, a comparison to gdb, and that's about it).

# ? Feb 20, 2016 00:15

the talent deficit: Dec 20, 2003; self-deprecation is a very british trait, and problems can arise when the british attempt to do so with a foreign culture

any serious software should have specification tests (given input a does this produce b?) and model tests (do the results of running random-but-valid operations against my thing match that of running the same operations against an abstract model of my thing?). anyone writing software where the cost of failure is high should read this cool and good paper: https://www.google.ca/url?q=http://...WDOXef21VDZLdqA

# ? Feb 20, 2016 04:29

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

some men just want to watch the world burn

https://github.com/citizenmatt/resharper-clippy

# ? Feb 20, 2016 11:00

Fanged Lawn Wormy: Jan 4, 2008; SQUEAK! SQUEAK! SQUEAK!

The other day I needed to do some debugging on my MIDI program I'm writing on a teensy

so I started just writing serial comments in that i could alternate comment out with the usbMIDI commands and in a few other small places just to check on if things did as expected

then, after upload, the controller would lock up and the VCP disappeared, making me enter 'programming mode' manually. eventually I get so frustrated I load up blink.ino just to see wtf
runs fine, VCP returns

start going through my code and commenting out changes, then realize I put a serial.print in the constructor of the custom piezotomidi object I created. the objects are declared before setup() even happens. sooooooo that explains the lockup

:whitewater:

PS: really need to use the debug/breakpoint features in visualmicro, haven't even tried them yet

# ? Feb 20, 2016 17:01

redleader: Aug 18, 2005; Engage according to operational parameters

gonadic io posted:

Jesus christ scala xml literals are implemented poorly. They seem fine, until suddenly you spend three hours working out why a test is failing on two identical xml nodes:
code:
val string = <Tag>{"1"}</Tag>
val int = <Tag>{1}</Tag>
val literal = <Tag>1</Tag>
 
println(string)
println(int)
println(literal)
println(string == literal)
println(int == literal)
Prints:
<Tag>1</Tag>
<Tag>1</Tag>
<Tag>1</Tag>
true
false

P.s. The explanation to this behaviour (until I knew exactly what to Google for) was in a comment on a stack overflow answer from 2011

i'll bite. why?

# ? Feb 20, 2016 23:54

Lutha Mahtin: Oct 10, 2010; Your brokebrain sin is absolved...go and shitpost no more!

NihilCredo posted:

some men just want to watch the world burn

https://github.com/citizenmatt/resharper-clippy

lol

# ? Feb 21, 2016 00:03

kitten emergency: Jan 13, 2008; get meow this wack-ass crystal prison

redleader posted:

i'll bite. why?

I think it's because you need a string inside the braces, but a bare 1 isn't a string

# ? Feb 21, 2016 00:05

akadajet: Sep 14, 2003

Lutha Mahtin posted:

lol

clippy powered ransomware

# ? Feb 21, 2016 00:07

gonadic io: Feb 16, 2011; >>=

redleader posted:

i'll bite. why?

the capability for tags to have non-string types is implemented as tag contents are parsed as Atom[_]s but this isn't actually exposed anywhere as pretty much nothing actually uses Atoms except for equality (due to equality first comparing types iirc).

i guess the intent was to allow type attributes etc?

so <Tag>1</Tag> and <Tag>{"1"}</Tag> are both Nodes that contain the normal Atom[String], but <Tag>{1}</Tag> is a Node that contains an Atom[Int].

code:

(<Tag>{"1"}</Tag>).child.head match {
    case atom : Atom[_] => println(atom.data.getClass)
}
(<Tag>{1}</Tag>).child.head match {
    case atom : Atom[_] => println(atom.data.getClass)
}

prints
class java.lang.Integer
class java.lang.String

you can't even match on Atom[Int] due to type erasure lol

the actual xml that they generate is the same though so any amount of checking the attributes or types or string representations or anything won't show the difference. all of the equalities (==, xml_==, strict_==) will fail though! so basically you're forced to 1) never insert any non-string into your xml, or 2) only compare using string equality (so it's fragile to ordering, open/close vs empty tags, whitespace, etc, etc)

martin odersky said that he didn't like the behaviour, that xml literals were a mistake, and that work needs to be done on them soon (in 2007). i guess most people just pretend that they don't exist.

gonadic io fucked around with this message at 00:27 on Feb 21, 2016

# ? Feb 21, 2016 00:22

fritz: Jul 26, 2003

Lutha Mahtin posted:

lol

WONTFIX. Working as intended.

# ? Feb 21, 2016 01:57

elite_garbage_man: Apr 3, 2010; I THINK THAT "PRIMA DONNA" IS "PRE-MADONNA". I MAY BE ILLITERATE.

freeing memory is whoopin' my rear end right now

# ? Feb 21, 2016 06:50

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

I keep getting true and false mixed up

# ? Feb 21, 2016 10:15

VikingofRock: Aug 24, 2008

Soricidus posted:

I keep getting true and false mixed up

Programming in Bash?

# ? Feb 21, 2016 18:49

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

nope, java. i just keep putting conditionals the wrong way round and then getting confused when the opposite thing happens to what i expected

it's always the little things

# ? Feb 21, 2016 19:22

Jerry Bindle: May 16, 2003

that happens to me, too, sometimes. i actually quite rarely use conditional statements that are more complex than if(state == Thing.STATE_A) though. using a < instead of a <= hosed me up for a moment last week.

# ? Feb 21, 2016 20:25

CPColin: Sep 9, 2003; Big ol' smile.

I always love when the planets align and something like this happens:

code:

if (!isNotAvailable())
{
   ...
}

# ? Feb 21, 2016 20:57

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

yeah, it's all about the negative verbs. i'm running into poo poo like "thing.setDisabled(thing.shouldBeEnabled)".

# ? Feb 21, 2016 21:01

Lutha Mahtin: Oct 10, 2010; Your brokebrain sin is absolved...go and shitpost no more!

that feeling in yr brain when you think it makes more sense to create a isNegative() function, but you know it's gonna mess you up later when you actually call it

# ? Feb 21, 2016 21:35

brap: Aug 23, 2004; Grimey Drawer

what about (< 0) is inadequate for you

# ? Feb 22, 2016 00:30

Fanged Lawn Wormy: Jan 4, 2008; SQUEAK! SQUEAK! SQUEAK!

Soricidus posted:

yeah, it's all about the negative verbs. i'm running into poo poo like "thing.setDisabled(thing.shouldBeEnabled)".

I need to reread smart dog thanks for triggering me

# ? Feb 22, 2016 02:03

VikingofRock: Aug 24, 2008

fleshweasel posted:

what about (< 0) is inadequate for you

Language without operator overloading?

# ? Feb 22, 2016 02:30

necrotic: Aug 2, 2005; I owe my brother big time for this!

why is it that !someNegative() has remained so prevalent? i just dont get the thought process that leads to that instead of creating a somePositive() to call instead.

# ? Feb 22, 2016 09:38

qntm: Jun 17, 2009

my favourite is when it turns out that the variable noRetries is not "disable retries" but actually "number of retries", or vice-versa

# ? Feb 22, 2016 10:32

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

necrotic posted:

why is it that !someNegative() has remained so prevalent? i just dont get the thought process that leads to that instead of creating a somePositive() to call instead.

it's because there are a lot of terrible programmers, op

# ? Feb 22, 2016 10:45

DaTroof: Nov 16, 2000; CC LIMERICK CONTEST GRAND CHAMPION
There once was a poster named Troof
Who was getting quite long in the toof

necrotic posted:

why is it that !someNegative() has remained so prevalent? i just dont get the thought process that leads to that instead of creating a somePositive() to call instead.

probably because a "negative" value like undefined/null/zero/false is commonly where a variable starts, and people wind up writing the function around that assumption out of habit

# ? Feb 22, 2016 10:47

DaTroof: Nov 16, 2000; CC LIMERICK CONTEST GRAND CHAMPION
There once was a poster named Troof
Who was getting quite long in the toof

or if they're using ruby they broke their skulls against poo poo like this

code:

class Foo
  def is_enabled
    @is_enabled ||= true
  end
  def disable
    @is_enabled = false
  end
end

foo = Foo.new
foo.is_enabled #=> true
foo.disable
foo.is_enabled #=> true

# ? Feb 22, 2016 10:54

Symbolic Butt: Mar 22, 2009; (_!_); Buglord

qntm posted:

my favourite is when it turns out that the variable noRetries is not "disable retries" but actually "number of retries", or vice-versa

this happened to me at least 3 times. before you said it it never occurred to me that the numero sign wasn't an exclusive thing to romance languages in this way

# ? Feb 22, 2016 10:57

qntm: Jun 17, 2009

no wait my favourite is Perl

Perl code:

unless(!$obj->retriesDisabled() == 0) {
    # what
} else {
    # WHAT
}

# ? Feb 22, 2016 11:05

ErIog: Jul 11, 2001

I have a crypto question. I worked on this today: http://cryptopals.com/sets/6/challenges/42/

I got it mostly working, but I'm using SHA256 so me finding a cube root with an algorithm I pulled off Rosetta Code didn't work. I can't find a cube root that leads to the proper and complete SHA256 hash. Most of the hash is there, but it kind of loses coherence past a certain point. I suspect that due to the length of SHA256 I'm making the problem harder than I need to, and it seems like SHA1 with the shorter ASN.1 and the shorter bit length would allow me to fit the data I need into the part of the number I have easier control over.

So my question is, is doing this exercise with SHA256 impossible? If it is possible, is it just a matter of me calculating the value of the garbage padding block by hand based on Finney's writeup of the exploit? Am I a terrible programmer for not being able to do that?

How is cube formed?

ErIog fucked around with this message at 11:32 on Feb 22, 2016

# ? Feb 22, 2016 11:29

suffix: Jul 27, 2013; Wheeee!

works for me with sha256

make sure you're using bigints or equivalent, floating point calculations will not be accurate enough

i binary search to find a number whose cube is anywhere in the range of numbers that starts with the shortest prefix + message hash

# ? Feb 22, 2016 12:30

jesus WEP: Oct 17, 2004

is it even possible to check a repo out for a vs project without fighting nuget for like 2 hours?

# ? Feb 22, 2016 16:35

jony neuemonic: Nov 13, 2009

qntm posted:

no wait my favourite is Perl
Perl code:
unless(!$obj->retriesDisabled() == 0) {
    # what
} else {
    # WHAT
}

to be fair, never using unless-else is about the only thing that entire community agrees on.

but... yeah. it's definitely special.

# ? Feb 22, 2016 16:48

DONT THREAD ON ME: Oct 1, 2002; by Nyc_Tattoo; Floss Finder

necrotic posted:

why is it that !someNegative() has remained so prevalent? i just dont get the thought process that leads to that instead of creating a somePositive() to call instead.

it's because someone else wrote the object you're calling somePositive on and you're not really responsible for maintaining that class so instead you write bad code

# ? Feb 22, 2016 17:18

computer toucher: Jan 8, 2012

just found out about python virtualenv and it's p cool.

look at me, I'm a baby learning to crawl.

# ? Feb 22, 2016 17:45

Bloody: Mar 3, 2013

St Evan Echoes posted:

is it even possible to check a repo out for a vs project without fighting nuget for like 2 hours?

hmm yup

# ? Feb 22, 2016 18:51

CRIP EATIN BREAD: Jun 24, 2002; Hey stop worrying bout my acting bitch, and worry about your WACK ass music. In the mean time... Eat a hot bowl of Dicks! Ice T; Soiled Meat

https://twitter.com/37point2/status/701817966678118400

# ? Feb 22, 2016 18:52

Shaggar: Apr 26, 2006

St Evan Echoes posted:

is it even possible to check a repo out for a vs project without fighting nuget for like 2 hours?

only if you setup the project correctly immediately after you created it. alternatively, the new project.json works way better now cause nuget is handled more consistently

# ? Feb 22, 2016 19:30

Adbot: ADBOT LOVES YOU

# ? May 11, 2024 16:21

gonadic io: Feb 16, 2011; >>=

More fun tales from dealing with 3rd party services

Me: this valid xml document is accepted but not all of the fields are obeyed
Rep: yeah, your field tags are in the wrong order in the address node.
Me: fair enough, that fixes it. Why does it silently ignore them rather than give a validation error? Do you have a schema so that this doesn't happen again?
Rep: yes: here's a sample request with some of the fields marked with . Note the ordering of the fields.

gonadic io fucked around with this message at 20:05 on Feb 22, 2016

# ? Feb 22, 2016 19:45

The Something Awful Forums > Discussion > Serious Hardware/Software Crap > YOSPOS > terrible programmers: my god, it's full of chars

«‹›1446 »