Is being annoyed/bad at C sign that I'm a bad coder or not fit for CS?

Haha I used to do the same thing. When you write programs try to find a reason to use pointers, like if you have to pass a large array into a function dereference it instead. Understanding pointers is best learned through a bit of practice, instead of trying to memorize w3schools. I spent years dabbling with C, but It took a week of trying to build my own game engine to get it all in 'muscle memory'

how the fuck do people not understand pointers?

Attached: 1541728454644.jpg (1280x720, 362K)

Did you write less than 100 lines of C total or are you a fucking brainlet? & is for getting a memory address, * is for pointer declaration and dereferencing pointers. C is not fucking C++ for fucks sake, it's a pretty simple language

Performance is the most unimportant thing you could possibly worry about for 99% of programming projects on your personal computer

why is python such a whitepsace nazi, it drives me crazy

I understand them but I hardly encounter uses in corporate jobs.

bitch you can't even get that far in the standard library without pointers

Another benefit to C and C++: Static executable with no dependencies, whilst you will need a Python interpreter installed on the machine to run your application.

They're used constantly. Even a language that doesn't have explicit pointers still uses them under the hood. A in a simple statement like

int x = 0;

x is actually an implicitly dereferenced pointer to some data. That data just happens to be on the stack instead of the heap. Pointers are pretty much everything unless you're using a language with more stack control like FORTH.

This
C is practically useless without pointers, you can't even use memory outside of whatever you have pre-allocated in the stack by your compiler.

Unfortunately not true. Its about 30%. The solution is to use python AND C.

>Is being annoyed/bad at C sign that I'm a bad coder or not fit for CS?

No, it's not. Many of the CS people I know (including myself) are annoyed by almost any imperative programming language. The language of CS isn't C but Abstraction.

>implying braindead CRUD web development for random medium sized businesses isn't 90% of development anyways

That list is dumb. Faster could mean a couple of minutes in C++ is a couple of days with python.

>C is a pain in the ass until it all clicks
it's not difficult to understand it's just tedious

using * and & as operators doesn't help

this

>*
pointy star

>&
and-dress

Python is a script language for script kiddies
C/C++is a programming language for software developers

Why does everyone say every fucking interpreted language is portable? You need to install the interpreter, that's a pain in the ass.
Java is another one of these: it's not portable, you need a fucking VM to make it run.

The only true portable languages are compiled ones. If you already need to write interpreters for every system why don't you make a compiler for it instead?

>C is not fucking C++
This is the problem. OOP fags are literally incapable of understanding anything that isn't obfuscated to hell and behind several layers of programmatic bureaucracy.

import btfo

def how
will(python[fags,ever]+'recover')
never = true

>It's all starting to come together for me except for pointers
So you don't know C then.

Python is a working languages for machine learning and AI purposes

>you can call it's functions from within any language
This. Most libs I use in phyton are actually written in C, because phyton is so fucking slow

>Import btfo
>Never use it
This is how fags like you write bloated software in a shit language

The best way to use C++ is C with STL, but you don't have to, you can throw in the fuck you want, it supports every paradigm.
Of course if you use it like Java you're a moron, and your brain doesn't work anymore because that language taught you even "main" should be an object.

Ok kid

If you ever do any systems programming, you'll learn to like C, or you'll become a fanboy of one of the new systems programming languages that actually stands a chance of competing with C (eg. Go, Rust,) not a language from a different domain that's designed to solve different problems (Python.) You'd also learn that C is more portable than Python, which is obvious, because the Python interpreter is written in C, so C can run on every platform that has Python, and on systems where Python isn't installed, and on systems that the Python interpreter isn't portable to. If you were going to design a new interpreted language like Python, which language would you write the interpreter in? C, probably, because C's portability would allow your language to run on a wide range of platforms.

That said, you should also learn that different languages are suited for different tasks. If you want to write a low-level utility or library, you probably want C, but if you want to do a lot of high-level stuff, like text processing, you're going to have a bad time in C, and you'd want to use something like Python.

I also think you're underestimating how much of programming is "really low level stuff" or requires better performance than Python can provide. Someone has to write the world's language interpreters, system tools, web browsers, video codecs, etc., and for the most part, they won't be written in Python.

Attached: 1537152271500.jpg (680x664, 67K)

>C isn't portable
>C is somehow larger and more complex than fucking Python when the entire language is described in seven easy to understand chapters
>hard to learn
>he's never read K&R C

>It's all starting to come together for me except for pointers
Pointers are literally the only parts of the language that need effort to learn tho.

not really with a mechanical keyboard
you start to enjoy all the typing

Because compiling implies compiling for a specific architecture. A compiled program will not necessarily run properly on different hardware, let alone on different OS. Java solves this with the universal JVM. Interpreted languages don't have this problem. You don't even know what you're arguing about.

People who learn languages like Python and C# have the behavior obfuscated. I know because I was one of them.

>saying "c/c++"
Spotting retards is so easy.
Also
>5-10 higher programmer productivity
lolno. Dynamically "typed" languages give you the illusion of productivity, since you need to write 5x more tests to replace what a good compiler wouls have done.
I have unironically had a faster time developing in C++ than Python. Many people say that dynamically "typed" languages save you development time, but in my experience, it's the opposite: statically typed languages actually save you time by preventing you from running code that is ill-formed anyway.

Wise user.

>since you need to write 5x more tests
you're missing the point

You should definitely learn C for the lessons in humility it provides. You must be smarter than your tools to use them correctly, and C will force you to understand everything. Taking automatic memory handling and high level data structures for granted without understanding them will make you a poor programmer. This field is not easy, and there are no shortcuts. Hit the books and learn the concepts from the bottom.

The top 0.01% of programmers routinely write memory handling bugs in C, and even after several rounds of peer review these bugs sometimes still make it into huge open source projects.
It's more than just a pain in the ass. It's programmer hostile. It very difficult to write actually correct C, and it's very easy to write C which merely appears to work. It's not a good language unless you absolutely need to use it.

Let's go through that slide

>Faster to learn

This is just barely true in the case of C. C++, this is absolutely true, but C is simple enough that it doesn't take much longer to learn than Python.

>Portable code

Portable C code is doable and really isn't very hard. Calling this a major benefit is an overstatement. Furthermore, when portability is enforced, it can create slower code, and makes it hard to create drivers or anything low-level in the language.

An example of something being rewritten in C from Python with the goal of increasing portability: gamingonlinux.com/articles/sc-controller-the-driver-and-ui-for-the-steam-controller-is-being-rewritten-to-be-more-portable.13040

>Much smaller and compact code

Only if you avoid high-level libraries. C code can be compact if you allow yourself to use the same level of libraries that you would in Python.

>Automatic garbage collection

Ok, but again, this isn't a major benefit

>Easy to find and install 3rd party libraries

This is actually a significant benefit, and although clibs.org/ attempts to help with it, it is still an issue.

>More powerful standard library

This is just a difference in philosophy. It isn't very hard to run `clib install buffer`. However, this is actually a valid benefit.

>5-10 times higher programmer productivity

The productivity increase doesn't come without disadvantages, and if you allow the use of libraries in C/C++, the gap decreases.

Nonetheless, this is still a benefit, which is why Python is a good prototyping and scripting language. However, using it for more than that is where the issues arise.

>Faster

Not only did you not create list of pros/cons instead of advantages for each, but you've completely ignored many C/C++ advantages, and understated 'faster'..

C code can, and often is, millions of times faster than Python. The reason for this is simple, for each line of Python that you write, thousands and millions of lines of C are executed. With a good selection of libraries, you can get similar productivity and much better speed.

In case anyone misses it though, all of the positives listed for Python are simply restatements of 'increases productivity'. Why not just leave it at that single point? Because this is quite literally Python propaganda.

>The top 0.01% of programmers routinely write memory handling bugs in C

This is actually a good point. Unlike the bullshit in the slide above.

>unless you absolutely need to use it
For example when you have to write actually safe software. Your only two options are C and Ada.

thanks for taking up 3/4 of my browser page with your beyond-reddit spacing

Sorry about that, it's a weird habit I have. I do it with my code too and people hate it.

If I had to read code like that I'd fire you

>5-10 times higher programmer productivity
Only for write once and never modify code. Evolving a complex project over time as requirements change requires either static typing or 5-10 times more debugging.

I'd hook up a dragon dildo to diff from clang-format. For every byte he'd get a minute of penetration.

No, it's not for everyone, wasn't for me either. It's not a blocker for your career either, just power through it and leave it if you're not interested in it beyond classwork.

>C and Ada
No. Just Ada. If you're writing safety critical software in C then you've already lost. Yes, I'm aware that it happens all the time anyway.

>However, using it for more than that is where the issues arise.
No, it's not. Having bad practices and terrible architecture in place is where things go wrong.

>If you're writing safety critical software in C then you've already lost
As demonstrated by the only safe kernel (down to binary) in existence being written in C.

Attached: 1531109129109.png (303x311, 178K)

seL4?

Know of any other safe kernel?

There was one obscure one that I stumbled upon that was at least aiming for similar verification, I can't seem to find it now though. I think it was also written in C though.

Python is handy when used to glue routines written in C/C++.

Also where is usual Python 2 vs Python 3 clash?

Attached: Еби_Чугун.jpg (341x356, 43K)

>Python is handy when used to glue routines written in C/C++.
You mean Lua.

Aiming for and being there are different things.
C is as viable as Ada for safe software.

Note 'at least'. I'm not sure how far it got

>how far it got
Not as far as sel4, since there's nothing on it.

Well if you're going to formally verify your software with external proofing tools then you could write your shit in anything. 99% of your work will be spent in external modelling anyway.

Daily reminder that the L4 was proofed against a design specification written in Haskell.

Go fuck yourself. You know for a fact that safe software where the budget is unlimited vs safe software in the real world are two separate things. Formal verification is exceptionally rare and takes space shuttle levels of manpower to complete. For everything else, there's languages where writing safe code without 5 man hours per line is possible. Do not pretend that you can write seL4 quality code in C. The C written in that project was practically manually assembled bytecode from a high level formal spec. C is as viable as Ada for safe software in the same way as directly wiring transistors together is viable for safe software. It's possible, but there's little reason to do it and achieving a good result will take far more time.

I build electromagnetics simulations for a living.
Fortran is my main language, but C is also fine.

Python is utterly slow. Numpy is very limited, so not really usable. Mathematica can do way more. It executes faster, faster to program and the library of special functions and plot routines and integration techniques is so far above python that it doesn't make sense.

To me programming in python is worst of both worlds. Slow as fuck, no reliable libraries. However, development time in python is almost the same as in a C or frotran.

Mathematica is also quite slow but development time is very short.

It seems to me that python is one of the reasons that we see so many bad programming today. It abstract many things and gives spergs the idea that they can program well. However, they still don't have to understand how a computer works, so they make all the wrong design decissions.

That being said, as a scripting language it's kind of nice. I think python is just an advanced bash alternative.

On the C/C++ side put down not a dynamically typed nightmare.

>python is just an advanced bash alternative.
Redpill of the day.

>Well if you're going to formally verify your software with external proofing tools then you could write your shit in anything.
Except you couldn't, because good luck proving anything in a nondeterministic environment.
>Daily reminder that the L4 was proofed against a design specification written in Haskell.
Your point? It's written in C which also has to be proven to implement the spec.
>You know for a fact that safe software where the budget is unlimited vs safe software in the real world are two separate things.
Budget is never unlimited. seL4 is a real-world project, so is JAS 39.
>For everything else, there's languages where writing safe code without 5 man hours per line is possible.
None of those are system languages. Ada without SPARK is almost as bad as C is. With SPARK it's about equal to C with FramaC. You still get the benefit of coq with FramaC, unlike the limited possibilities of SPARK.
>Do not pretend that you can write seL4 quality code in C. The C written in that project was practically manually assembled bytecode from a high level formal spec.
Do not pretend you can write seL4 quality code in Ada. The toolchain they developed for seL4 is, without any doubt, the most astonishing practical result in formal verification. They chose C for a reason, you can read all about that in their earlier papers.
>It's possible, but there's little reason to do it and achieving a good result will take far more time.
Ada and C are on equal footing when it comes to safety-critical (not even verified, where C is a much better choice to target) ever since FramaC.
Plain C is worse than plain Ada, sure, but who uses those in safety-critical... It's not the 90's anymore.

I work in data science. Even here, everything production-facing is Scala or Java. Python is just used for throw away prototypes but even then, anyone with a PhD uses R.
I think Python is the most overrated shit of all time.

>The top 0.01% of programmers routinely write memory handling bugs in C, and even after several rounds of peer review these bugs sometimes still make it into huge open source projects.
>It's more than just a pain in the ass. It's programmer hostile. It very difficult to write actually correct C, and it's very easy to write C which merely appears to work. It's not a good language unless you absolutely need to use it.

That's the same for all languages that are this close to hardware. I don't think fortran is much better. Assembler is of course way worse.

Higher level languages are of course safer. However, they are also much more bloated. Software written in C or the likes is so fast that it will be close to THE best possible solution. In bloat languages you end up rewriting the same thing. One way of programming is good for one application but bad for another. For example because you have to initialize bloat many times.

Added tot that. Numerical thing in Python are mostly just interfaces to things written in C/fortran.

>Automatic garbage collection
This is a bad thing.
>Portable
False

Why does everyone say every fucking compiled language is portable? You need to install the compiler, it's a pain in the ass. C is another one of these: it's not portable, you need a fucking compiler for the specific architecture and OS to make it compile (+adapt your code because everyone platform has its own quirks. just look at the shitstorm in autotoole needed to compile for every platform).

The only true portable language is assembly. If you already need to write a compiler for every system why don't you make an assambler for it instead?

> Faster
Apache is easier to learn, for example, but everybody use Nginx, because it's faster. Now, if we're speaking about clients - nobody cares about these plebs and Python is acceptable.

Please pay attention to class next time.

Do you even know what assembly is? Do you know what instruction sets are?

Don't school me with your basic shit, boy. I was highlighting his ridiculous argument.

You use pointers in nearly everything... references are just pointers in Java. Same thing in python and javascript. There's some constraints on what you can do with them and some additional syntactic sugar but they're implemented as pointers.

Is the point writing shit that crashes several times a day?

>gamingonlinux.com/articles/sc-controller-the-driver-and-ui-for-the-steam-controller-is-being-rewritten-to-be-more-portable.13040
fucking hilarious that they are switching from python to c for
>muh portability
and then decide to use GTK instead of Qt

>learns to randomly switch * and & until something works
>gets employed
>code now used in xray machines
>gives kids cancer
learn what you are doing before you hurt someone!

* - value
& - address
not hard stuff.

What operators would you choose?

Sure, I don't know what I'm talking about.
If you think the JVM doesn't need to be compiled for every architecture as well I don't know how you think computers work.

Maybe C isn't portable, but it doesn't need 2 programs to run one.

So, with what do you implement the JVM then?

Nothing is really portable if you dig deep enough you fucking retards.

Can you guys please recommend some Python books?

are you literally or actually dumb? please chose your proficiency.

C is a brainlet filter. If you can't master something this simple, you have no business in computer science.

this

Yes.

>Also where is usual Python 2 vs Python 3 clash?
Normally in print statements, python 2 print was a statement, in 3 it's a function. There are a few others, but in my experience that's the most common issue.

>you'd want to use something like python
No, you'd want to use a saner scripting language, like lua or bash/grep/sed/awk. Python is unbearably slow for anything outside of programming lessons and the only reason anyone should use it is when they absolutely need the libraries it has

>Python books

lmgtfy.com/?q=python+books

>You need to install the compiler, it's a pain in the ass. C is another one of these: it's not portable, you need a fucking compiler for the specific architecture and OS to make it compile

And you need to do the same for something like Python. You need the fucking interpreter, which originally needs to be compiled on a compiler. Get it now, dipshit?

No, nobody is going to do anything in assembly, because C is not like assembly. Your attempt at mocking him is retarded.

Attached: python.jpg (884x1325, 119K)

cool, I'm a network architect so I'm actually just a scripter.

Enjoy your pajeets.