|
HoboMan posted:maybe where you live is poo poo because i had at least one a day maybe? city of 400k? anyway can't move. might just have to start my own pharmacy
|
# ? Aug 21, 2016 21:54 |
|
|
# ? May 25, 2024 15:05 |
|
hobbesmaster posted:agree on 2 but for 1 on linked in I'm almost 100% certain their queries include your skills. ie my title is software engineer but I don't get front end spam maybe but probably not job title is everything and "software engineer" is one of the titles for a front-end gig. so there you have it
|
# ? Aug 21, 2016 22:40 |
|
what is it for back-end then? because gently caress front-end
|
# ? Aug 22, 2016 02:03 |
|
HoboMan posted:what is it for back-end then? because gently caress front-end it's all the same job titles are meaningless
|
# ? Aug 22, 2016 02:11 |
|
echinopsis posted:there is just so few jobs. maybe one pops up every week or two? have you ever thought of becoming an internationally renowned gigolo?
|
# ? Aug 22, 2016 07:30 |
|
has anyone said gigolo economy yet
|
# ? Aug 22, 2016 11:45 |
|
echi get a magento certification and become my code monkey at least I can talk to you in english and you'll understand me, this is a huge plus
|
# ? Aug 22, 2016 14:26 |
|
Symbolic Butt posted:magento certification
|
# ? Aug 22, 2016 14:32 |
|
lol I know you never know how desperate a dude is
|
# ? Aug 22, 2016 14:36 |
|
3rd party API's error logs are abbreviated to like 1000 characters lol they be like "An error occured, Java.Lang.NullPointerException in com.3rdparty.ButtFunc(str (...)" HoboMan fucked around with this message at 15:47 on Aug 22, 2016 |
# ? Aug 22, 2016 15:11 |
|
if you're ever leading a c++ project, please for the love of god enforce const-correctness
|
# ? Aug 22, 2016 16:39 |
|
const_cast #yolo
|
# ? Aug 22, 2016 16:47 |
|
quiggy posted:if you're ever leading a c++ project, please for the love of god enforce const-correctness in college we used c++ for half of our comp sci classes that required programming, which led me to the false impression that i knew how to write c++. i'm pretty sure we never covered the concept of const-correctness in any of those classes.
|
# ? Aug 22, 2016 16:50 |
|
same
|
# ? Aug 22, 2016 16:55 |
|
hobbesmaster posted:const_cast i'd literally prefer code peppered with reinterpret_casts and void pointers than const_cast holy gently caress what a terrible idea
|
# ? Aug 22, 2016 17:01 |
|
const_cast and reinterpret_cast are there because casts to (void*) are legal so you may as well let people describe exactly what retarded thing they're doing
|
# ? Aug 22, 2016 17:02 |
|
GameCube posted:in college we used c++ for half of our comp sci classes that required programming, which led me to the false impression that i knew how to write c++. i'm pretty sure we never covered the concept of const-correctness in any of those classes. this infuriates me. it's not like const-correctness is a hard concept, it's just something to be aware of. just tag your const methods const, people!
|
# ? Aug 22, 2016 17:16 |
|
have you ever seen code written by college professors
|
# ? Aug 22, 2016 17:20 |
|
quiggy posted:this infuriates me. it's not like const-correctness is a hard concept, it's just something to be aware of. just tag your const methods const, people! I always forget that const is a thing
|
# ? Aug 22, 2016 17:21 |
|
void * is not a foolproof "raw pointer" value in C++ either, due to the possibility of implicit conversion operators. the official raw pointer type is const volatile char& (actually a reference type, so to cast a pointer you have to dereference it first e: I'm an idiot, of course if you have a pointer you can just cast it directly because they are built-in types and you can't give them a user-defined conversion operator), as seen for example in std::addressof
hackbunny fucked around with this message at 23:02 on Aug 22, 2016 |
# ? Aug 22, 2016 17:22 |
|
quiggy posted:this infuriates me. it's not like const-correctness is a hard concept, it's just something to be aware of. just tag your const methods const, people! *reading visual studio express 2005 build output, in idiot undergrad voice* cannot assign to a variable that is const? what the gently caress? *selects the word "const" by holding trackpad button with one finger and dragging over it with the other, code compiles* wow computers are loving bullshit, why would they include a word that just makes your code not compile *submits homework assignment, gets wasted at 6 pm on a tuesday*
|
# ? Aug 22, 2016 17:26 |
|
hackbunny posted:const volatile
|
# ? Aug 22, 2016 17:27 |
|
uncurable mlady posted:recruiters: they're bad
|
# ? Aug 22, 2016 17:35 |
|
and you thought that "when is a const not a const" quip was a joke a "volatile" variable can have a different value every time it's read, so every time you fetch a "volatile" variable, the compiler has to insert a memory fetch (not a memory fence though: it may - Microsoft's compiler does - but it's not required and you shouldn't expect it). basically it's like a variable with a getter and a setter function (literally sometimes, if they're used to describe memory mapped hardware registers for example), although compilers will often try to pretend they don't have to be that strict... and get it wrong. I forget who was that professor working on program correctness checking, but he showed that clang had bugs in its treatment of volatiles by modifying it to explicitly use functions to read/write volatiles, which resulted in stricter and more correct behavior than the default implementation a "const volatile" variable is a variable with a getter but no setter. the getter still has to be called every time, hence volatile I believe that with memory mapped registers you totally can have write-only variables but C and C++ have no way to represent that
|
# ? Aug 22, 2016 17:39 |
|
that's so the compiler doesn't whine if the thing you're taking the address of is volatile or const
|
# ? Aug 22, 2016 17:39 |
|
yeah in that context it's just the most permissive set of pointer/reference qualifications available
|
# ? Aug 22, 2016 17:41 |
|
ya in embedded land write-only addresses are very common or, better yet, they'll pair a read-only register with a write-only register and just stick them at the same address, so the read-contents and write-contents are different and at best vaguely related
|
# ? Aug 22, 2016 17:47 |
|
a bug worthy of the underhanded C contest: an audit chastised us (actually a third party library) for using rand, so I replaced it with (int)arc4random_uniform(UINT_MAX) this makes rand return negative numbers about half the time, meaning things like rand() % 100 (lovely way to get a uniform distribution, I know, but not my code) can now return values out of range (from -99 to 99 instead of from 0 to 99) which breaks all sorts of things... randomly the fix is one byte: change UINT_MAX to INT_MAX
|
# ? Aug 22, 2016 17:58 |
|
hackbunny posted:I forget who was that professor working on program correctness checking, but he showed that clang had bugs in its treatment of volatiles by modifying it to explicitly use functions to read/write volatiles, which resulted in stricter and more correct behavior than the default implementation http://blog.regehr.org
|
# ? Aug 22, 2016 18:00 |
|
fritz posted:probably :
|
# ? Aug 22, 2016 18:09 |
|
hackbunny posted:the fix is one byte: change UINT_MAX to INT_MAX christ I'm an idiot, it's (unsigned)INT_MAX + 1
|
# ? Aug 22, 2016 18:16 |
|
GameCube posted:*reading visual studio express 2005 build output, in idiot undergrad voice* cannot assign to a variable that is const? what the gently caress? *selects the word "const" by holding trackpad button with one finger and dragging over it with the other, code compiles* wow computers are loving bullshit, why would they include a word that just makes your code not compile *submits homework assignment, gets wasted at 6 pm on a tuesday* they're right though, const is almost entirely worthless
|
# ? Aug 22, 2016 18:35 |
|
proper use of it can prevent deep copies at least
|
# ? Aug 22, 2016 18:38 |
|
hackbunny posted:(int)arc4random_uniform(UINT_MAX) i'm the casting of uint to int
|
# ? Aug 22, 2016 18:45 |
|
gonadic io posted:i'm the casting of uint to int should it be (int)(unsigned)? idk, now that the range is right it shouldn't matter
|
# ? Aug 22, 2016 18:46 |
|
"it's changed by something else, so do a real memory read every time you need it. also, i'm making a promise that I won't write to it in the space of this program"
|
# ? Aug 22, 2016 19:20 |
|
hackbunny posted:a "const volatile" variable is a variable with a getter but no setter. the getter still has to be called every time, hence volatile i learned something new today! most likely utterly useless to me since i'm not a hardware guy, but who knows with c++
|
# ? Aug 22, 2016 19:32 |
|
loving owned for like 5 hours in stage because i didn't know java has its own CA trust process
|
# ? Aug 22, 2016 19:48 |
|
So the architect likes "just hash it" to the point he wants to rip out all the stupid search loops. We walked through the code together (tribal knowledge, woo!! gently caress writing poo poo down!) and he kept apologizing for how bad the code was. He's juggling 3 projects and didn't really have fresh in his mind how bad what we took on to fix actually is. When I ask about biz rules with $client they'll admit they don't know, or that some definitions are nonstandard. I thought consulting was glamorous not "oh my gooooood why didn't you call us sooner." On the flip side I feel better about my code.
|
# ? Aug 22, 2016 20:31 |
|
|
# ? May 25, 2024 15:05 |
|
Symbolic Butt posted:echi get a magento certification and become my code monkey what is this? something forgeniers get to remote work from their hut in mongolia
|
# ? Aug 22, 2016 20:42 |