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 »

bawk: Mar 31, 2013

# ? Apr 12, 2016 01:42

Adbot: ADBOT LOVES YOU

# ? May 24, 2024 22:45

bawk: Mar 31, 2013

god loving dammit i hate this stupid pre-baked SALR shameful snipe poo poo, sorry for swiping what could have been a good snype because i misclicked on the "SNYPE" button instead of the unread reply count, everyone

# ? Apr 12, 2016 01:44

JewKiller 3000: Nov 28, 2006; by Lowtax

delete all encodings and normalization forms from the unicode standard except utf-8 and nfc

# ? Apr 12, 2016 01:44

compuserved: Mar 20, 2006; Nap Ghost

qntm posted:

I thought Unix filenames were opaque sequences of bytes, so this would be a standpoint consistent with that

this doesn't work well when git claims to be cross-platform and other operating systems / filesystems actually enforce a filename encoding and normalization form (mac os x). it's neither cool nor good when you commit a file with filename "y�sp�s" whose pathname is UTF-8 encoded and someone whose default encoding is UTF-16LE (for the sake of argument) clones your repo and sees "潹灳獯" instead. hopefully not even git is that dumb but i wouldn't be surprised.

also, reminder that mac os x is a certified unix and they're able to enforce a pathname encoding just fine

JewKiller 3000 posted:

delete all encodings and normalization forms from the unicode standard except utf-8 and nfc

agreed but maybe nfd. or UTF-16. who cares. just standardize on something

compuserved fucked around with this message at 02:43 on Apr 12, 2016

# ? Apr 12, 2016 01:49

compuserved: Mar 20, 2006; Nap Ghost

tef posted:

git config --global core.precomposeunicode true

cool git thanks for making users deal with this nonsense (although it's really linux's fault more than anything else's for not enforcing a filename encoding in the first place)

compuserved fucked around with this message at 02:03 on Apr 12, 2016

# ? Apr 12, 2016 01:59

Bloody: Mar 3, 2013

HoboMan posted:

this is the first thing i looked at after google "tef blog": http://programmingisterrible.com/
i don't know if it's his, but it looks like it belongs here

yes that is correct

# ? Apr 12, 2016 02:45

Bloody: Mar 3, 2013

my build process for a visual Studio project amounts to a Jenkins scripted msbuild that dumps its artifacts in a directory that gets committed to svn which is how we release software

# ? Apr 12, 2016 02:46

JimboMaloi: Oct 10, 2007

HoboMan posted:

this is the first thing i looked at after google "tef blog": http://programmingisterrible.com/
i don't know if it's his, but it looks like it belongs here

thats it and it is great. mononcqcs is http://ferd.ca and is also great.

e: oh poo poo, accidental https shaming

JimboMaloi fucked around with this message at 02:56 on Apr 12, 2016

# ? Apr 12, 2016 02:46

Bloody: Mar 3, 2013

quote:

Your connection is not secure

The owner of ferd.ca has configured their website improperly. To protect your information from being stolen, Firefox has not connected to this website.

zoinks

# ? Apr 12, 2016 02:53

JimboMaloi: Oct 10, 2007

HoboMan posted:

so what the gently caress is DevOps?

depends who you ask. if a job listing says "devops engineer" it probably means "youre in ops but we expect you to automate poo poo".

in theory the term was meant to describe a culture where devs and ops actually spoke to each other instead of throwing binaries over the wall, but that takes effort and its much easier to coopt the term and use it to stroke the egos of people who view ops as a lesser position (see also 'reliability engineer' as its used at startups)

# ? Apr 12, 2016 02:54

Bloody: Mar 3, 2013

HoboMan posted:

so what the gently caress is DevOps?

http://danluu.com/google-sre-book/

# ? Apr 12, 2016 02:56

anthonypants: May 6, 2007; by Nyc_Tattoo; Dinosaur Gum

JimboMaloi posted:

depends who you ask. if a job listing says "devops engineer" it probably means "youre in ops but we expect you to automate poo poo".

in theory the term was meant to describe a culture where devs and ops actually spoke to each other instead of throwing binaries over the wall, but that takes effort and its much easier to coopt the term and use it to stroke the egos of people who view ops as a lesser position (see also 'reliability engineer' as its used at startups)

at my last job the way they described it was application support for datacenter customers but the way it was implemented was having the engineers try to automate azure server deployments

# ? Apr 12, 2016 02:58

MeruFM: Jul 27, 2010

i'm more confused about Continual Delivery Engineer. I thought continuous delivery was a process encompassing stuff like CI, "agile", and "forever updated" applications like chrome.

# ? Apr 12, 2016 03:00

Bloody: Mar 3, 2013

i interpret continuous delivery as being a push to master is a push to prod

# ? Apr 12, 2016 03:02

Luigi Thirty: Apr 30, 2006; Emergency confection port.

xamarin doesn't seem to be terrible for writing a toy app

# ? Apr 12, 2016 03:13

VikingofRock: Aug 24, 2008

Notorious QIG posted:

hey fun c++03 fact: if you have a std::vector<Object*> v, then v[i]+1 compiles fine and without warning!

Really? I get a -Wunused-value warning, which is part of -Wall. (Link to example)

Unless you mean you were using that value somewhere in a surprising way, which is an understandable complaint but which is pretty reasonable behavior given how pointer arithmetic works in C++. Image that you are actually storing arrays of Objects in your vector--you would want v[i]+1 to compile in that case so that you can index your array. Like this:

C++ code:

#include <iostream>
#include <vector>
#include <string>

using namespace std;

int main(){
    string arr[3];
    arr[0] = "yos";
    arr[1] = "pos";
    arr[2] = "bithcoin";

    vector<string*> v;
    v.push_back(arr);
    cout << "*(v[0]+1): " << *(v[0] + 1) << endl; // prints the element at v[0][1], which is "pos";
}

# ? Apr 12, 2016 03:17

quiggy: Aug 7, 2010; [in Russian] Oof.

VikingofRock posted:

Really? I get a -Wunused-value warning, which is part of -Wall. (Link to example)

Unless you mean you were using that value somewhere in a surprising way, which is an understandable complaint but which is pretty reasonable behavior given how pointer arithmetic works in C++. Image that you are actually storing arrays of Objects in your vector--you would want v[i]+1 to compile in that case so that you can index your array. Like this:
C++ code:
#include <iostream>
#include <vector>
#include <string>

using namespace std;

int main(){
    string arr[3];
    arr[0] = "yos";
    arr[1] = "pos";
    arr[2] = "bithcoin";

    vector<string*> v;
    v.push_back(arr);
    cout << "*(v[0]+1): " << *(v[0] + 1) << endl; // prints the element at v[0][1], which is "pos";
}

unless im mistaken you're only getting unused value if you don't actually then read it. also the only reason that you get "pos" in your example is that you already put "yos" and "pos" in an array so they're guaranteed to be adjacent, if they're not already adjacent you'll segfault when you try to access v[0] + 1

# ? Apr 12, 2016 03:21

Bloody: Mar 3, 2013

I thought vector guaranteed adjacency

# ? Apr 12, 2016 03:24

quiggy: Aug 7, 2010; [in Russian] Oof.

Bloody posted:

I thought vector guaranteed adjacency

vector guarantees adjacency, but the vector type is Object*. the Objects themselves are in no way guaranteed to be adjacent since it entirely defends on how and where they were defined

as a really lovely example that you should never actually write:

C++ code:

int a = 3;
int* b = new int;
std::vector<int*> v;
v.push_back(&a);
v.push_back(b);

this should compile and run fine, even though v[0] is a pointer to an int on the stack and v[1] is a pointer to an int on the heap

# ? Apr 12, 2016 03:26

Progressive JPEG: Feb 19, 2003

JimboMaloi posted:

(see also 'reliability engineer' as its used at startups)

fiber is my reliability engineer

# ? Apr 12, 2016 03:39

Gul Banana: Nov 28, 2003

git filenames are a huge pain on windows also
it will happily add both path/thing.c and path/thing.C to the repo. guess what happens when you check that out on NTFS

go on, guess. because I certainly don't know, except that it's bad

# ? Apr 12, 2016 04:29

JewKiller 3000: Nov 28, 2006; by Lowtax

that wouldn't be very good on most mac os x hfs+ filesystems either (which are case-insensitive by default)

# ? Apr 12, 2016 04:35

JawnV6: Jul 4, 2004; So hot ...