This is not meant to be inflammatory or anything like that, but I am in the midst of learning C, and (think) I have a good handle on most of the basics. I've done all of the various book exercises: primes generators, Fibonacci generators, string manipulation, yadda yadda, but none of this is cool.
What is the "bridge" between command line programs and something -cool-? I've heard of various games being written in C, but how?
Forgive my exasperation, but it feels like I've been learning lots but can still only do relatively little. Thanks for any insight on what to do with C.
Relevant information: OS X leopard, PHP and web development experience (which is so great because projects are immediately in a context where you recognize how they can be powerful)
C is the concrete and the steel of modern tech
There was a time when almost everything was written in C, or in something much worse.
These days, many of the advanced languages and systems are actually implemented in C or C++, and then those things implement more systems. It is standing on the shoulders of giants, as the expression goes. Almost every OS kernel, browser, and heavy-duty-web-server is written in C/C++.
So sure, you don't see the steel in the high rise, you see the beautiful interior furnishings and the sleek glass windows. You don't want a steel or concrete desk, and if you did, it would be too expensive to build for you.
Back to your GUI question: your first C graphics program should probably use the original X Window System directly. Don't spend too much time there, but then move on to one of the more advanced Widget toolkits such as GTK+ or (the C++) Qt. Be sure to investigate your OS X system, as it has one of the most advanced of them all.
I try love to write things in Ruby these days, but I happen to know there are over 100,000 200,000 lines of C code implementing that cool Ruby language system. :-)
Summary
Ok, this post got really big, so here's a quick summary before you read it: to program GUIs, there are a number of good C/C++ libraries (for example, QT). What most people do, on Windows systems at least, is to use a .NET language (like C#) when they want GUIs, and C/C++ when they want more control/speed. It's also very common to use both in combination, i.e. make a GUI in C# and speed-critical parts in C.
I still encourage you to read the longer answer, it contains a lot more information on your options.
Longer answer
I'll start with the big question, then answer (as best I can) your specific question about creating a GUI. I think you're kind of suffering from the fact that C is used to teach programming, and it's much easier to do so only using command line programs (after all, they're much simpler to write). This doesn't mean that C can't do all of the stuff you want, like GUIs specifically. It does. I don't think there's any type of software that hasn't been done in C, usually before it was done in other languages.
All right, some answers:
Is C Useful?
C, and its very close relative C++, are responsible for a huge portion of the world's code. I don't know if more code is written in C than any other language, but I'm guessing it's not far off.
Most of the really important programs you use are actually written in C/C++. Just for one example, Windows.
Where is C used today?
C/C++ are still used a lot. They're especially useful for developing low-level stuff (i.e. stuff like Operating Systems, which need a lot of speed, a lot of ability to control exactly what your code does, etc.).
But don't think it's all low level for C programmers. Even today, with many other languages available (which are arguably much better, and certainly much easier to program), C is still used to create practically everything. GUI applications, which you specifically asked about, are very often made with C, even though nowadays, lots of people are switching to other languages. Note I say switching: C used to be the standard language for writing, well, everything, really.
How do I develop GUIs with C
Alright, you specifically wanted to know how to create a GUI with C (I'm hoping C++ is ok too).
First of all, it depends on a number of factors:
What Operating System are you writing for? (Windows, Mac, and Linux are the most common).
Do you want the GUI to work on other systems as well?
The most common case is writing software to work on Windows. In that case, the "natural" solution is to write things that work with the Win32 API. That's basically the library that "comes" with Windows, letting you do any GUI work you want to do.
The big problem with this is, it's kinda hard. As in, a lot hard. This is the reason most people don't do that kind of work anymore.
So what are your other options?
The most natural is going with what's called a .NET language. Those are a bunch of languages, together with libraries, that Microsoft created. They're probably the easiest way to get a GUI on Windows. The problem is, you can't really use them from C (since it's not a .NET language).
Assuming you want to stay in C/C++ land, you can use some kind of library which makes working with the Windows API easier (since it hides all the ugly details):
One of the most common is what's called MFC (Microsoft Foundation Classes), which are a bunch of C++ classes which make it "easier" to create Windows stuff. Unfortunately, this library is very old, and is really not that easy to use.
The other way to go if you want to program in C++ is to use some kind of third-party library. This is a library that someone other than Microsoft created, which makes it easier to create a GUI.
Another option is to create only the GUI part of your software in a .NET language, and use C/C++ for the other parts (or use the .NET language to do almost everything, and use C/C++ only when you really need it, for example when you need things to go really fast). This is a very popular option.
The advantage of a third party library is that, if you pick the right one, you can use the same code to create a GUI for all the Operating Systems at the same time.
I think the most famous of these libraries is QT, and also WxWidgets, but there are a bunch of other ones. I would personally pick QT since it seems to get the most fame, but I haven't worked with either.
Every major operating system has all of its low-level libraries implemented in C. Mac OS X is a Unix-like system under the hood, which is a wonderful world if you're a C programmer.
Check out The Art of Unix Programming for some great ideas.
As for GUI stuff, you'll probably want to use X11. There is plenty of good documentation out there -- most Unix programming stuff and most deep system-level stuff just assumes you're working in C, since that's what everybody uses for it.
Well, that depends. If you want to build desktop applications, a multiplatform GUI library whose main language is C is GTK+:
http://www.gtk.org/
For games, check out SDL:
http://www.libsdl.org/
Which provides you with thinks like direct input from keyboard, 2D graphics, some audio and even threads and stuff like that. It can also open an OpenGL context if you want to get into the 3D world (however it's hard to do it in raw OpenGl). Did I mention that SDL is multiplatform?
However the real strength of C is in systems programming. For desktop applications/games maybe you are best suited with C++. Now that you have command of C, learning the basic C++ should be easy ;).
Cool stuff do with C?
Operating Systems, device drivers, and python modules for starters.
Games typically will use C++ if they're C-Based, in my experience / what I've seen.
There are many libraries for C under Unix, such as X lib, which accesses X11.
If you wanted to get into robotics you may find C to be very useful, as you will have to write low-level code with very small memory footprints, so even C++ may not be the best choice.
C and C++ are very good at writing small, fast code, but OOP is not always the best choice, so at times you will find that C is a better choice, for example if you are writing a compiler or OS.
Sure there are some impressive programs made in C !
GNOME for example, arguably the most used desktop environment used in modern unix systems is written in C (the major parts at least) and is mostly based on GTK+ gui toolkit, itself done in C.
For game, OpenGL is a C api and is the standard for 3D graphic programming in multi-platform development (not uniquely microsoft platform), and Quake 3, which the engine, Id Tech 3, is available in GPL, is also writen in C. There also is many 2D games written using SDL library.
SDL is a good library for graphics and sound, and I've seen some cool stuff done with it. If you do it in C, it'll take longer to make, but from a performance point of view, it'll be much better.
If your idea of cool is GUI apps and you want to write native GUI apps on the Mac, you'll want to look at
Carbon. This is the official C API into the Mac GUI and OS. They keep threatening to kill it, but it survives.
Personally I think GUI apps are a very narrow definition of cool. What I think is cool is implementing parallelized math algorithms using opencl.
GTK-server is REALLY easy to get started with, in C or any other language. Just click that link.
For a "cool" application that goes beyond simple GUI's, check out the OpenCV computer vision library. It provides fast real-time image processing and face recognition.
Now you can access a webcam and start writing real-time computer vision games. For applications like these that are processor intensive, C is the ideal choice.
Last I checked more Open Source projects are started in C than in any other language.
The fact that C is used by so many large and successful projects doesn't particularly make it "good". The reason C is so commonly used is because of a few factors, it's been around a long time, it's fast, it lets you access both low level and high level interfaces as needed, and it's better than the other old languages (FORTRAN etc). The "cool" thing about C is that you can make it do absolutely anything: inject itself into the kernel and add some new features or bug fixes that you couldn't convince Microsoft to do, etc.
Yes, C can be and is easily used for things beyond the command line, but it's extremely dangerous due to pointers... Not to mention development in other more modern languages is faster (and safer) by magnitudes. I never use C unless it's the last resort, ie: need to implement something low level or needs that extra performance.
By the way, when I say C, I really mean C++. I'd never choose C over C++ unless I was forced to.
Related
I want to write a library that is dynamically loadable and callable from C code, but I really don't want to write it in C - the code is security critical, so I want a language that makes it easier to have confidence that my code is correct. What are my options?
To be more specific, I want C programmers to be able to #include this, and -l that, and start using my library just as if I had written it in C. I'd like programmers in other languages to be able to use their favourite tools for linking to C libraries to link to it. Ideally I'd like that to be possible on every platform that supports C, but I'll settle for Linux, Windows and MacOS.
Anything that compiles to native code. So you might Google for that - "languages that compile to native code." See, e.g., Programming languages that compile to native code and have the batteries included
C++ is often the choice for this. Compiles to native code and provided you keep your interfaces simple, easy to write an adapter layer.
Objective C and Fortran are also possible.
It sounds like you are looking for a language with ABI compatibility or which can be described as resulting in native code. So long as it can be compiled to a valid object file (typically an .obj or .o file) which is accepted by the linker, that should be the main criteria. You also then want to write a header file as a convenience for any client code which is written in C (or a closely related language/variant thereof).
As mentioned by others, you need a pretty good reason for choosing a language other than C as it is the lingua-franco of low-level/systems software. Assembler is an option, although harder to port between platforms. D is a more portable - but less widespread - alternative which is designed to produce secure, efficient native code with a minimum of fuss. There are many others.
Almost every security critical application I know of is written in C. I don't believe that there are any other language that has higher real status in producing secure applications.
C is being said to be a poor language for security by people who don't understand.
If you want C programmers to use your library, use C. Doing anything else is tying one hand behind your back whilst trying to walk on a balance beam (the gymnastics equipment). Sure, there are dozens of other languages that are CAPABLE of interfacing to C, but it typically involves using a C layer and then stuffing the C data types into a language specific data type (Java Objects, Python Objects, etc, etc), and when the call is finished, you use the same conversion back to a C data type. Just makes it harder to work with, and potentially slower if you don't get all the design decisions right. And people won't understand the source code, so won't like to use it (see more about this below).
If you want security, then write very good code, wearing your "security aspects" hat firmly on at all times, find a security mailing list or website and post it there for review, take the review comments on board, understand the comments, and fix any comments that are meaningful to fix. Distribute the source code to the users, so people can see what your code does. Those that understand security will know what to look for and understand that you have done a good job (or a bad job, whichever is applicable) - and those who don't will hopefully trust the right pople. If it's good, people will use it. If it's "hidden", and not easy to access, you won't get many customers, no matter what language you use.
Don't worry, you won't reveal anything more from releasing source. If there is a flaw in the code, and it is popular (or important) enough, someone will find the flaw, even if you publish only binaries. For those skilled in reverse engineering, not having source code is only a small obstacle.
Security doesn't stem from using a specific language or a specific tool, it stems from good design and good basic understanding of the problems with security.
And remember security by obscurity (whether that means "hidden source code" or "unusual language" or something else obscure) is false security.
You might be interested in ATS, http://ats-lang.sourceforge.net/. ATS compiles via C, can be as efficient as C, and can be used in a way that is ABI-compatible with C. From the project website:
ATS is a statically typed programming language that unifies implementation with formal specification. It is equipped with a highly expressive type system rooted in the framework Applied Type System, which gives the language its name. In particular, both dependent types and linear types are available in ATS. The current implementation of ATS (ATS/Anairiats) is written in ATS itself. It can be as efficient as C/C++ (see The Computer Language Benchmarks Game for concrete evidence) and supports a variety of programming paradigms
ATS's dependent and linear type system helps produce static guarantees about your code, including various aspects of resource management safety.
Chris Double has been writing a series of articles exploring the power of ATS's type system for systems programming here: http://bluishcoder.co.nz/tags/ats/. Of particular note is this article: http://bluishcoder.co.nz/2012/08/30/safer-handling-of-c-memory-in-ats.html
This document covers aspects of calling back and forth between ATS and C code: https://docs.google.com/document/d/1W6DYQApEqKgyBzMbvpCI87DBfLdNAQ3E60u1hUiMoU0
The main downside is that dependently-typed programming is still a daunting prospect, even for non-systems programming. The syntax of the language is also a bit weird: consider lexical quirks such as the use of abst#ype as a keyword. Finally, ATS is to some degree a research project, and I personally don't know whether it would be sensible to adopt for a commercial endeavour.
Theoretically, it's going to be Fortran: less indirection (as in: my array is [here], not just a pointer to here, and this is true of most but not all of your data structures and variables).
However... There are many gotchas and quirks in Fortran: not, perhaps, as many as in C but you probably know your way around C rather better than Fortran. Which is the point behind most of the comments saying 'Know your code' - but do you really know what your compiler is doing?
Knowing you, I'm prepared to take it on trust that you do, for C. Most programers don't. You do not know and cannot know what a local JVM or JIT compiler does, and that's a black hole in your security model if you're using Java or C# r scripting languages.
Ignore anyone who tells you that the hairy-chested he-men of secure computing write their own assembler: they probably don't even know the security errors they're making in any and all nontrivial projects they release. Know your compiler, indeed.
You could write it in lua - providing a C API to a Lua library is relatively straight forward. C++ is also an option, though of course you'd have to write C wrappers and make sure no exceptions can escape your functions. But honestly, if it's security critical the minor inconveniences of the C language shouldn't be that much of a big deal. What you really should be doing is prove the correctness of your program where feasible, and test extensively where it's not.
You can write a library in Java. JNI is normally used to call C from Java, but it can be used the other way around.
There is finally a decent answer to this question: Rust.
I'm looking for a higher-level system language, if possible, suitable for formal verification, that compiles to standard C, so that it can be run cross-platform with (relatively) low overhead.
The two most promising such languages I've stumbled during the past few days are:
BitC - While the design goals of this language match my needs (it even supports the functional paradigm), it is in very unstable state, the documentation is out of date, and, generally, it seems like a very long shot for a real-world project.
Lisaac - It supports Design-by-contract, which is very cool and has a relatively low performance overhead. However the website is dead, there hasn't been a new release since '08 and generally it seems the language is dead.
I'd also like to note that it's not meant for a real-time system, so a GC or, generally, non-determinism (in the real-time sense), is not an issue.
The project involves mainly audio processing, though it has to be cross-platform.
I assume someone would point me to the obvious answer - "plain ol' C". While it is truly cross-platform and very effective, the code quantity would probably be greater.
EDIT: I should clarify that I mean cross-platform AND cross-architecture. That is why I consider only languages, compiled to C in the first place, but if you can point me to another example, I'd be grateful :)
I think you may become interested in ATS. It compiles to C (actually it expresses and explains many C idioms and patterns from a formal type-theoretic perspective, it has even been proposed to prepare a book of sorts to show this -- if only we had more time...).
The project involves mainly audio processing, though it has to be cross-platform.
I don't know much about audio processing, I've been mainly doing some computer graphics stuff (mostly the basic things, just to try it out).
Also, I am not sure if ATS works on Windows (never tried that).
(Disclaimer: I've been working with ATS for some time. It is bulky and large language, and sometimes hard to use, but I very much liked the quality of programs I've been able to produce with it, for instance, see TEST subdirectory in GLES2 bindings for some realistic programs)
The following doesn't strictly adhere to the requirements but I'd like to mention it anyway and it is too long for a comment:
Pypy's RPython can be translated to C. Here's a nice talk about it. It's been used to implement Smalltalk, JavaScript, Io, Scheme, Gameboy (with various degree of completeness), but you can write standalone programs in it. It is known mainly for its implementation of Python language that runs on Intel x86 (IA-32) and x86_64 platforms.
The translation process requires a capable C compiler. The toolchain provides means to infer various things about the code (used by the translation process itself) that you might repurpose for formal verification.
If you know both Python and C you could use cython that translates Python-like syntax to C. It is used to write CPython extensions.
I know very little about OOP, so maybe my question is silly, but still....
Can you access object-oriented (OO) API from procedural (non-OO) languages? For example, Win32 API is not OO, but I know there is a wrapper for C++ to make it OO. But is it possible to do it both ways?
I ask because I don't like OO languages; I learned C by programming microcontrollers, and OO languages just take the actual code away from you, and I worry that OOP is so popular that soon everything will become based on objects.
Let me approach this as one cranky old programmer from the 70's to another(?)
One common technique in the old days when we were writing a library was to have an "init" call that created some kind of "cookie" (usually a pointer or array index). Then we'd force the client to supply the cookie back to us on every other call to the library. That allowed us to do whatever bookeeping our library required (in a re-entrant manner) without bugging the client with all the implementation details. As a C programmer you should be very familiar with this style, because C uses it for all its file I/O, as did Unix. Microsoft liked to call them "handles" instead of "cookies". Unix called them "files" or sometimes "file handles".
A large amount of what OO languages do is to just add some extra syntax around this technique. Now instead of all your calls starting with LibnameCallname (cookie, ... they start with cookie.callname(.... But really, in a lot of ways this is just a syntax change to make things easier for us programmers (saves us typing that extra unneeded Libname on everything).
Now in a way, OSes (including Unix, which used the file as its basic paradigmn for damn near everything), are already OO, and have been OO for decades. How they handle OS calls is really just a linkage detail. It doesn't matter much to us as systems programmers, except that the linkages have to match. The only real problem with calling C++ from C isn't that its "OO". The problem is that C++ uses some custom name-mangling algorithm for its symbols, which not all C compilers can deal with.
In truth, there's a bit more to it than that. However, if someone were to write all their new OS calls in C++, you can bet the C compiler vendors would find a way to bridge the gap, so that you could call them in your comfy old LibnameCallname(cookie,... style from C.
What you are actually saying you don't like it seems to me is what Parnas first referred to back in 1972 as Information Hiding - The idea that developers can be more productive if the details about how the various parts of a program work are hidden from the other parts.
It was very contraversial at the time. Even as late as the early ninties I used to hear big arguments about it. Fredrick Brooks even argued against it in The Mythical Man-Month. (BTW: If you haven't read TMM-M, you need to.)
The thing is, almost nobody argues against it today. There is a reason for that. Even Fred Brooks admitted he was wrong twenty years later. From his esay titled, Parnas was Right, and I was Wrong about Information Hiding -
Parnas was right, and I was wrong. I
am now convinced that information
hiding, today often embodied in
object-oriented programming, is the
only way of raising the level of
software design.
To be fair, both sides would agree that on very small systems (like, for instance, the embedded systems you have been playing with?) information hiding is not nearly as nessecary. It is only when systems start getting really large that a fully interconnected system will start falling down under its own weight. However, today most programs are that large. That's the main reason the arguments have ceased.
Once the platform APIs become object-oriented, most likely there will be no choice of how to use them - only OO-based languages will be available. Eg. Android (Java only), Windows Phone 7 (C# only), WebOS and Blackberry (Java, JavaScript).
In general, it's possible (but very non-trivial) to reach classes and objects from plain C, but, as said, C just won't be an option on that platforms. This becomes a problem already, when portable C code can't be used on the above platforms (luckily, Android offers NDK for native development).
After five years of professional Java (and to a lesser extent, Python) programming and slowly feeling my computer science education slip away, I decided I wanted to broaden my horizons / general usefulness to the world and do something that feels more (to me) like I really have an influence over the machine. I chose to learn C and Unix programming since I feel like that is where many of the most interesting problems are.
My end goal is to be able to do this professionally, if for no other reason than the fact that I have to spend 40-50 hours per week on work that pays the bills, so it may as well also be the type of coding I want to get better at. Of course, you don't get hired to do things you haven't done before, so for now I am ramping up on my own.
To this end, I started with K&R, which was a great resource in part due to the exercises spread throughout each chapter. After that I moved on to Computer Systems: A Programmer's Perspective, followed by ten chapters of Advanced Programming in the Unix Environment. When I am done with this book, I will read Unix Network Programming.
What I'm missing in the Stevens books is the lack of programming problems; they mainly document functionality and provide examples, with a few end-of-chapter questions following. I feel that I would benefit much more from being challenged to use the knowledge in each chapter a la K&R. I could write some test program for each function, but this is a less desirable method as (1) I would probably be less motivated than if I were rising to some external challenge, and (2) I will naturally only think to use the function in the ways that have already occurred to me.
So, I'd like to get some recommendations on how to practice. Obviously, my first choice would be to find some resource that has Unix programming challenges. I have also considered finding and attempting to contribute to some open source C project, but this is a bit daunting as there would be some overhead in learning to use the software, then learning the codebase. The only open-source C project I can think of that I use regularly is Python, and I'm not sure how easy that would be to get started on.
That said, I'm open to all kinds of suggestions as there are likely things I haven't even thought of.
Reinvent a lot of the core Unix utilities. Most of these were (and still are) written in C, so they are a good way to start off learning. Depending on your skill, pick harder or easier utilities to copy.
Try writing your own malloc. You'll learn a lot about Unix and a lot of C programming as well.
Google for computer science operating system courses and do the projects there. Many schools have these projects on public websites so you could get everything you need. Here is a link to Purdue's site. Give the shell project a shot; it was difficult, but really educational.
Here are a few stackoverflow postings discussing C/Unix programming books. Their main claim to fame is extensive fan-out to other resources.
Practice some of the idioms (understand the ins and outs of pointers etc) and pick projects that help with that. The third item in the list has an answer (disclaimer, I wrote it) with a bunch of high-level C idioms where an idiomatic C program would differ from Java.
Learning XLib or Win32 GUI programming with C is probably less useful as almost anything is better than C for GUI programming and wrapping a core C engine with another language is generally much easier - unless you're really concerned with speed. As a starting point, concentrate on 'systems' programming applications, which is where you are most likely to get real mileage from C. Learn how the C interface of one of the scripting languages like Tcl or Python works.
Writing some bit-twiddly and pointer heavy code (e.g. picking apart network packets or interpreting a protocol), or server-side items using sockets will take you right into C's core competencies. If you've got some of the WRS books, try making pthreads work (IIRC UNP2 has a big section about pThreads). For a contrast, write something that uses non-blocking I/O. This stuff is C's home turf and you'll get a good working knowledge of C by doing this. Troll through the RFCs for some ideas of network protocols to implement.
What are good Linux/Unix books for an advancing user?
What are some good resources for learning C beyond K&R
Resources for learning C program design
What’s a good way to start learning about Data Structures & Algorithms?
Algorithms in C
Are you open to book suggestions? Although it is a bit dated (where "a bit" perhaps is a huge understatement), Maurice Bach's "The Design of the Unix Operating System" is a great book. It belongs next to K&R on your bookshelf.
You might try working your way through all the examples in the book Software Tools (Amazon). Much of it is pretty pedestrian (right-justify text, de-tabify, etc.), but it's a great introduction to the Unix philosophy and basic C programming.
This is pretty interesting. As we know that Unix is designed by using the C language, it may not difficult to learn. If you are using the Windows OS, you can use a "Windows services for Unix" application to practice your programs. If you are using Unix, then we can use editors like vi.
I would recommend one thing highly.
Try and re-write all usual Linux command lines tools like ls, mkdir, cd etc.
You'll gain a lot of knowlege about programming and Linux both at the same time.
Pick the commands from the easiest, say "time" and work all the way up to the more complicated ones.
The best way to consolidate your learnings it to practise. So just choose a kind of application that interest you and start developing it (for example, a network client/server simple application).
Try to test most of the Unix APIs (files, sockets, etc.) to see how they work. You could for example get an Unix programming book, follow its chapters and test on your application everything you read, by creating your own functions. On that way, you can start to develop your own function library to be used in further projects.
Write a webserver.
Make it multi-threaded.
Have it support a new scripting language you develop (a la PHP, etc.)
Allow uploads from authenticated users.
Write a plugin for your favorite tool (i.e. integrate with SVN to give a webview).
Closed. This question is opinion-based. It is not currently accepting answers.
Want to improve this question? Update the question so it can be answered with facts and citations by editing this post.
Closed 7 years ago.
Improve this question
Joel and company expound on the virtues of learning C and how the best way to learn a language is to actually write programs using that use it. To that effect, which types of applications are most suitable to the C programming language?
Edit:
I am looking for what C is good at. This likely does not coincide with the best way of learning C.
Code where you need absolute control over memory management. Code where you need to be utterly in control of speed versus memory trade-offs. Very low-level file manipulation (such as access to the raw devices).
Examples include OS kernel, and embedded apps.
In the late 1980s, I was head of the maintenance team on a C system that was more than a million lines of code. It did database access (Oracle), X Windows graphics, interprocess communications, all sorts of good stuff. It ran on VMS and several varieties of Unix. But if I were to recreate that system today, I sure wouldn't use C, I'd use Java. Others would probably use C#.
Low level functions such as OS kernel and drivers. For those, C is unbeatable.
You can use C to write anything. It is a very general purpose language. After doing so for a little while you will understand why there are other "higher level" languages available.
"Learn C", by all means, but don't don't stick with it for too long. Other languages are far more productive.
I think the gist of the people who say you need to learn C is so that you understand the relationship between high level code and the machine internals and what exaclty happens with bits, bytes, program execution, etc.
Learn that, and then move on.
Those 100 lines of python code that were accounting for 80% of your execution time.
Small apps that don't have a UI, especially when you're trying to learn.
Edit: After thinking a little more on this, I'd add the following: if you already know a higher-level language and you're trying to learn more about C, a good route may be to not create a whole new C app, but instead create a C DLL and add functions to it that you can call from the higher language. In this way you can replace simple functions that your high language already has to ensure that you C function does what it should (gives you pre-built testing), lets you code mostly in what you're familiar with, uses the language in a problem you're already familiar with, and teaches you about interop.
Anything where you think of using assembly.
Number crunching (for example, libraries to be used at a higher level from some other language like Python).
Embedded systems programming.
A lot of people are saying OS kernel and device drivers which are, of course, good applications for C. But C is also useful for writing any performance critical applications that need to use every bit of performance the hardware is capable of.
I'm thinking of applications like database management systems (mySQL, Oracle, SQL Server), web servers (apache, IIS), or even we browsers (look at the way chrome was written).
You can do so many optimizations in C that are just impossible in languages that run in virtual machines like Java or .NET. For instance, databases and servers support many simultaneous users and need to scale very well. A database may need to share data structures between multiple users (threads/processes), but do so in a way that efficiently uses CPU caches. In C, you can use an operating system call to determine the size of the cache, and then align a data structure appropriately to the cache line so that the line does not "ping pong" between caches when multiple threads access adjacent, but unrelated data (so called "false sharing). This is one example. There are many others.
A bootloader. Some assembly also required, which is actually very nice..
Where you feel the need for 100% control over your program.
This is often the case in lower layer OS stuff like device drivers,
or real embedded devices based on MCU:s etc etc (all this and other is already mentioned above)
But please note that C is a mature language that has been around for many years
and will be around for many more years,
it has many really good debugging tools and still a huge number off developers that use it.
(It probably has lost a lot to more trendy languages, but it is still huge)
All its strengths and weaknesses are well know, the language will probably not change any more.
So there are not much room for surprises...
This also means that it would probably be a good choice if you have a application with a long expected life cycle.
/Johan
Anything where you need a minimum of "magic" and need the computer to do exactly what you tell it to, no more and no less. Anything where you don't trust the "magic" of garbage collection to handle your memory because it might not be as efficient as what you can hand-code. Anything where you don't trust the "magic" of built-in arrays, strings, etc. not to waste too much space. Anything where you want to be able to reason about exactly what ASM instructions the compiler will emit for a given piece of code.
In other words, not too much in the real world. Most things would benefit more from higher level abstraction than from this kind of control. However, OS code, device drivers, and a few things that have to be near optimal in both space and speed might make sense to write in C. Higher level languages can do pretty well competing with C on speed, but not necessarily on space.
Embedded stuff, where memory-usage and cpu-speed matters.
The interrupt handler part of an OS (and maybe two or three more functions in it).
Even if some of you will now start to bash heavily on me now:
I dont think that any decent app should be written in C - it is way too error prone.
(and yes, I do know what I am talking about, having written an awful lot of code in C myself (OSes, compilers, VMs, network protocols, RT-control stuff etc.).
Using a high level language makes you so much more productive. Speed is typically gained by keeping the 10-90 rule in mind: 90% of CPU time is spent in 10% of your code (which you should optimize first).
Also, using the right algorithm might give more performance than optimizing bits in C. And doing things right in C is so much more trouble.
PS: I do really mean what I wrote in the second sentence above; you can write a complete high performance OS in a high level language like Lisp or Smalltalk, with only a few minor parts in C. Think how the 70's Lisp machines would fly on todays hardware...
Garbage collectors!
Also, simply programs whose primary job is to make operating-system calls. For example, I need to write a short C program called timeout that
Takes a command line as argument, with a number of seconds for that command to run
Forks two child processes, one to run the command and one to sleep for N seconds
When the first of the child processes exits, kills the other, then exits
The effect will be to run a command with a limit on wall-clock time.
I and others on this forum have tried several different solutions using shells and/or perl. All are convoluted and none quite do the right thing. In C the solution will be easy, because all the OS facilities are right where you can get at them.
A few kinds that I can think of:
Systems programming that directly uses Unix/Linux or Win32 system calls
Performance-critical code that doesn't have much string manipulation in it (e.g., number crunching)
Embedded programming or other applications that are severely resource-constrained
Basically, C gives you portable, efficient access to the native capabilities of the machine; everything else is your responsibility. In particular, string manipulation in C is tedious, error-prone, and generally nasty; the most effective way to do extensive string operations with C may be to use it to implement a language that handles strings better...
examples are: embedded apps, kernel code, drivers, raw sockets.
However if you find yourself more productive in C then go ahead and build whatever you wish. Use the language as a tool to get your problem solved.
c compiler
Researches in maths and physics. There are probably two alternatives: C and C++, but such features of the latter as encapsulation and inheritance are not very useful there. One could prefer to use C++ "as a better C" or just stay with C.
Well most people are suggesting system programming related things like OS Kernels , Device Drivers etc. These are difficult and Time consuming. Maybe the most fun thing to with C is console programming. Have you heard of the HAM SDK? It is a complete software development kit for the Nintendo GBA , and making games for it is fun. There is also the CC65 Compiler which supports NES Programming (Althought Not Completely). You can also make good Emulators. Trust Me , C is pretty helpful. I was originally a Python fan, and hated C because it was complex. But after yuoget used to it, you can do anything with C. Now I use CPython to embed Python in my C Programs(if needed) and code mostly in C.
C is also great for portability , There is a C Compiler for like every OS and Almost Every Console And Mobile Device. Ive even seen one that supports some calculators!
Well, if you want to learn C and have some fun at the same time, might I suggest obtaining NXC and a Lego Mindstorms set? NXC is a C compiler for the Lego Mindstorms.
Another advantage of this approach is that you can compare the effort to program the Mindstorms "brick" with C and then try LEJOS and see how it goes with Java.
All great fun.
Implicit in your question is the assumption that a 'high-level' language like Python or Perl (or Java or ...) is fast enough, small enough, ... enough for most applications. This is of course true for most apps and some choice X of language. Given that, your language of choice almost certainly has a foreign function interface (FFI). Since you're looking to learn C, create a module in the FFI built in C.
For example, let's assume that your tool of choice is Python. Reimplement a subset of Numpy in C. Since C is a pretty fast language, and has, in C99, a clear numeric library interface, you'll get the opportunity to experience the power of C in an appropriate setting.
ISAPI filters for Internet Information Server.
Before actually write C code, i would suggest first read good C code.
Choose subject you want to concentrate on, basically any application can be written in C, but i assume GUI application will be not your first choice, and find few open source projects to look into.
Not any open source project is best code to look. I assume that after you will select a subject there is a place for another question, ask for best open source project in the field.
Play with it, understand how it's working modify some functionality...
Best way to learn is learn from somebody good.
Photoshop plugin filters. Lots and lots of interesting graphical manipulation you can do with those and you'll need pure C to do it in.
For instance that's a gateway to fractal generation, fourier transforms, fuzzy algorithms etc etc. Really anything that can manipulate image/color data and give interesting results
Don't treat C as a beefed up assembler. I've done some serious app's in it when it was the natural language (e.g., the target machine was a Solaris box with X11).
Write something with some meat on it. Write a client server chess program, where the AI is on a server and the UI is displaying in X11; once you've done that you will really know C.
I wonder why nobody stated the obvious:
Web applications.
Any place where the underlying libraries are entirely in C is a good candidate for staying in C - openGL, Lua extensions, PHP extensions, old-school windows.h, etc.
I prefer C for anything like parsing, code generation - anything that doesn't need a lot of data structure (read OOP). It's library footprint is very light, because class libraries are non-existent. I realize I'm unusual in this, but just as lots of things are "considered harmful", I try to have/use as little data structure as possible.
Following on from what someone else said. C seems a good language to implement the language in which you write the rest of your software.
And (mutatis mutandis) the virtual machine which runs the rest of your software.
I'd say that with the minuscule runtime environment and it's self-contained nature, you might start by creating some CLI utilities such as grep or tail (or half the commands in Unix). Anything that uses only STDOUT, STDIN and file manipulation is a good candidate.
This isn't exactly answering your question because I wouldn't actually CHOOSE to use C in such an app, but I hope it's answering the question you meant to ask--"what would be a good type of app to use learn C on?"
C isn't actually that bad a language--it's pretty easily to understand your code at an assembly language level which is quite useful, and the language constructs are few, leaving a very sparse language.
To answer your actual question, the very best use of C is the one for which it was created--porting itself (and UNIX) to other CPU architectures. This explains the lack of language features very well; it also explains the existence of Pointers which are really just a construct to make the compiler work less--any code created with pointers could be created without it (just as well optimized), but it becomes much harder to create a compiler to do so.
digital signal processing, all Pure Data extensions are written in C, this can be done in other languages also but has had good success in C