/* Here I sit, Joe broken hearted, came to do some sh*t, but only just started. */

关于一些繁重的正则表达式输入验证。

int Q13Factor = 8125; // 2^13 for Q13 

首先，在一个大对象的更新的开始:

/*General note to all who tread in the <ObjectName>() code...
 * The SetOriginals() method from the BaseEntity class should be called (and only called) right after the Get() method
 * call as seen above.  Calling the SetOriginals method elsewhere will result in bugs and all kinds of other nasty suprises.
 */

然后在大约200行逻辑之后更新对象:

//Attempt to explain this confusing mess of code:
//First time you save an actual absence this is what happens:
//0. The first save saves to the <TableName> table (among other things). (Fig. A)
//1. The <CalculationMethod> method is called next which inserts to the <OtherTableName> table. 
//(This is the table that keeps track of credits to the case.) (Fig. B)
//2. So then you have to call <UpdateCalculations> to move the <TableName> records to the <ThirdTableName> table. (Fig. C)
//3. Then you go back and run calculations since you have the debits table (<ThirdTableName>) populated. (Fig D.)
//4. Then a final save to save the calculations back to the case. (Fig. E)
//Yeah, I know what you're thinking:  this sucks. 10/01/07 XXX

开发商是对的……这太糟糕了!

Many years ago (about 1994) I was working on a Oracle PRO*C application for a large multi-national software company that you will have heard of. The app I was working on was a massive Oracle application and they had a utility that ran overnight tidying up data and doing all sorts of aggregate calculations. Every time anything needed doing as a batch job, it got shoved into this utility and as you can imagine it became an absolute monstrosity. It was also notable for the tiny number of comments that it had for such a massive program.

它为数不多的评论之一仍然是我所见过的对纯粹的WTF的最好的评论……我试图在一个函数中找到一个bug，它有数百行长，中间是函数中唯一的注释:

/* I did this the other way */

直到今天，这仍然是我所见过的最好的评论。

在头文件的顶部:

/* Project : XYZ (Please somebody shoot me!)
 *
 * File : $Id: defs.h,v 1.1 $
 *
 * Purpose : Create havoc rather than peace among many nations
 *
 * History : Back-ported changes that were not in CVS.  Please somebody,
 *  shoot us and put us all out of our misery.
 */

“XYZ项目”(已改名)是一个长达七年的折磨。最后这句话是由一个从头到尾都参与其中的坚定的灵魂写的。

在linux 1.0内核调度器(schedule .c)中:

Dijkstra可能讨厌我。

/*
 *  'schedule()' is the scheduler function. It's a very simple and nice
 * scheduler: it's not perfect, but certainly works for most things.
 * The one thing you might take a look at is the signal-handler code here.
 *
 *   NOTE!!  Task 0 is the 'idle' task, which gets called when no other
 * tasks can run. It can not be killed, and it cannot sleep. The 'state'
 * information in task[0] is never used.
 *
 * The "confuse_gcc" goto is used only to get better assembly code..
 * Dijkstra probably hates me.
 */
asmlinkage void schedule(void)

(…)