I encourage everyone to have a glance over the Quake II source code

[Vespasian_2]

Better yet, any source code that was released by ID software.

You can find it all, here:

https://github.com/id-Software/Quake-2

It is so incredibly clean, elegant and pushes everything out of the C programming language including some Godlike pointer hacking and bit field manipulation. Function pointers and callbacks are made to look smooth.

This is the work of the programming God John Carmack and his team Brian Hook and George Cash.

The opinions are not new, this article outlines the beauty of Doom III's source code:
The Exceptional Beauty of Doom 3's Source Code

These guys were magicians and I aspire to be like them one day.

[Elysia]

Sounds like a nightmare to me.

[stevesmithx]

My props to John Carmack and his team. And I like the fanboy in you, but this raises a question: If John is God, then what do you call Dennis Ritchie? Meta-God??

[Vespasian_2]

My props to John Carmack and his team. And I like the fanboy in you, but this raises a question: If John is God, then what do you call Dennis Ritchie? Meta-God??

Yep, Dennis Ritchie and Kernighan Meta Gods

[Elkvis]

Yep, Dennis Ritchie and Kernighan Meta Gods

And Ken Thompson and Bjarne Stroustrup.

[brewbuck]

I've looked over this code before. It's... eh... decent, I guess. There are plenty of tricks and hacks that make the code fast enough to pull off a really awesome video game -- and I don't dispute whatsoever that Quake II was an awesome game -- but it's hardly a shining example of clean, well-architected code.

It got the job done and made Carmack and his coworkers a lot of money. That's all well and good for Carmack, but you're not Carmack. You should set your sights a little higher if you're looking for examples of well-designed code.

[Yarin]

but it's hardly a shining example of clean, well-architected code

What code is?

[Elkvis]

What code is?

Windows 2000

[Yarin]

Windows 2000

I forgot how good this was. They really comment well.

// HACK ON TOP OF HACK ALERT

A paragon of code quality if I don't say so myself.

[brewbuck]

What code is?

The source code of Clang.

[Mario F.]

The source code of Clang.

That was funny. Now, an application that isn't obviously a C programming tool.

I think you are being too hard on Mr. Carmack and the OP. From constraints in the development time to the mediocre skill level of a large portion of professional developers, it's always a reason to rejoice when we can find generally smart and clean production code that also respects best practices in such a complex language like C or C++.

[brewbuck]

That was funny. Now, an application that isn't obviously a C programming tool.

I think you are being too hard on Mr. Carmack and the OP. From constraints in the development time to the mediocre skill level of a large portion of professional developers, it's always a reason to rejoice when we can find generally smart and clean production code that also respects best practices in such a complex language like C or C++.

I'm just not moved to tears by the use of a function pointer...

My biggest problem with this code is the complete lack of tests. I mean, not a single test to be found anywhere. It also has "break" and "continue" statements all over the place. If it was code I was reviewing at work, I'd reject it for either of those reasons.

Again, I'm not saying the code sucks, or Carmack is stupid, or anything like that. I just don't see anything super impressive here.

[Epy]

The source code of Clang.

Which means nothing, because Clang is nothing but a glorified debugger. Anyone who has tried to use Clang for "real" code, should know what I'm referring to, i.e. Clang producing ud2 instructions unnecessarily, without warning.

Edit: RANT: anyone who tries to defend Clang here is automatically wrong. This has been an issue for over 3 years now, and has not been fixed.

[brewbuck]

Which means nothing, because Clang is nothing but a glorified debugger. Anyone who has tried to use Clang for "real" code, should know what I'm referring to, i.e. Clang producing ud2 instructions unnecessarily, without warning.

Edit: RANT: anyone who tries to defend Clang here is automatically wrong. This has been an issue for over 3 years now, and has not been fixed.

Your pet frustration notwithstanding, clang is pretty much taking over everything now. Android and iOS are both compiled with it, and Microsoft is switching to it. Get over it, man.

[Elysia]

Then enlighten us as to the problem because from what I see the problem is very vague.

From what I gather, you're complaining about how clang generates these trap instructions for undefined behaviour because they are messing up your code. And as the article says, they're generated because the compiler cannot always catch undefined behaviour at compile time. Are you trying to imply there's a bug that causes these instructions to trap when there's actually no undefined behaviour at all?