It's always great when a good book is released under a free license. This should be the normal case for all books that are out of print.

Unfortunately, most books simply vanish and/or have an unclear legal status.

A quick question: the book (at least early on) advocates this for main:

    # implied int; implied void
    main() {
        # whatever
        exit(0) # or EXIT_SUCCESS
    }

Is this simply a stylistic choice (rather than a more explicit int main(void) and return(0)). I'm mostly curious, but I know that how to deal with main can be a sacred cow.

Edit: And sure enough clang (by default) and gcc (with -Wall) both issue warnings.

The book was written 20 years ago, before the C99 spec was released. main() was acceptable but discouraged in C89, which is why -Wall (or more precisely -Wimplicit-int) displays a warning. It's disallowed by the C99 spec and will always give you a warning.

The OP's link looks like an HTML dump of http://publications.gbdirect.co.uk/c_book/ , which I'm pretty sure is the original...

(And there's also a PDF of all the pages, linked to from the gbdirect version, http://publications.gbdirect.co.uk/c_book/the_c_book.pdf ; for those of you having trouble with the printer-friendly version.)

Awesome timing. I'm just starting to contemplate porting the prototype VM i'm working on to C. Thanks Mahmud! You seem to be coming to the rescue often these days :-p

The book is from 1991, almost 20 years ago. There is a lot of outdated information. You should perhaps look for something more recent, that covers C99 and features commonly found on current compilers.

I was given this book for my 15th birthday - it rescued me from Sinclair Basic! It may be old but it's a very readable C tutorial and won't lead people far astray.

Thanks for the heads-up. Is C really that different now that I need to worry about it? I have done some work in C++ before, but that was many years ago and I've grown accustomed to high-level languages...

In essence, it is the same language, just that it evolved considerably since then. There are places where you could really use a more current reference. For example, the book instructs you to define the main function like:

    main(){
      /* ... */
    }

This is considered bad code style for todays standards. It doesn't state the return value (it defaults to int) and it leaves the argument list open (you can pass any number of arguments to the function).

In other places, it puts restrictions that are no longer relevant today. For example, since C99 (and most compilers allowed this long before C99) you can declare variables anywhere inside a block (which is pretty handy), and not just at the beginning of the function.

Its the same, but C99 adds a couple of neat features like boolean types, variable length arrays, variadic macros, inline functions and a lot more.

No it's not that different. GCC still compiles 1978 C.

Good thing I'm using GCC then! Thanks Mahmud!

What's up with the printer friendly version? It doesn't work for me.

Good to know it's not just me. (Chrome dev 6.0.458.1 and Safari both no-go on OSX 10.6.4.)

Do you have no feelings at all for your fellow man? :(

Misery loves company? More seriously, I take your point. I say that a lot ("Good to know it's not just me."). I'll have to think a little bit more about why.

I was only kidding Shared joy is a double joy; shared sorrow is half a sorrow

> I say that a lot ("Good to know it's not just me."). I'll have to think a little bit more about why.

I say it a lot too, especially when talking about interfacing with external resources, and I don't think it's anything to do with enjoying others' misery. It's simply being happy to have it confirmed that the problem you're observing is (probably) someone else's fault, not your own; this carries the bad news that you (probably) can't fix it yourself, but the good news that you don't have to fix it yourself.

Here's the official GNU C book: http://www.htu.tugraz.at/~past/ctut/ctut.pdf