|
You're right there. But as KarstenK said, deep nesting is an indication that the code is too complex.
There are very few loops out there that cant be broken down... sure you may need to use functions to accomplish it sometimes, but the cleaner code is well worth that effort.
Bikram Singh
|
|
|
|
|
I wasn't implying that I agreed with deeply nested conditionals and loops! Quite the opposite. Just observing that they're very common. This is another way of saying that there are lots of poor programmers out there.
Kevin
|
|
|
|
|
I know !
The implication that i implicated was that the code was too complex, which implied that the implication i implicated was correct. I didnt imply that i was implicating you for your note, but just that i didnt really mean to implicate what you thought i was implicating.
Bikram Singh
|
|
|
|
|
Just use goto and if some dumbass starts asking questions, just quote Stroustrup (TC++PL 3rd ed).
My programming blahblahblah blog. If you ever find anything useful here, please let me know to remove it.
|
|
|
|
|
Nemanja Trifunovic wrote:
just quote Stroustrup (TC++PL 3rd ed).
For those of us that do not have this book, would you please provide the relevant quote?
"When I was born I was so surprised that I didn't talk for a year and a half." - Gracie Allen
|
|
|
|
|
If you don't have it, you really should... Still there are those who can't afford it (yet...).
This is from the special edition, which should be identical to the 3rd edition except most of the errata is corrected. (I used snip where I cut some text that is interesting, but doesn't apply here) I copied this by hand, so mistakes are mine.
Page 137:
The goto has few uses in general high-level programing, (snip)
The goto can also be important in the rare save in which optimal efficiency is essential, (snip)
(snip a couple paragraphs)
One of the few sensible uses of goto in ordinary code is to break out from a nested loop or switch statement.
There you have it: Most of the time goto should not be used. However there are exceptions to that rule, where goto makes the code more readable, or where there is critical speed problem that can be corrected by goto, use it.
Remember when applying the latter that the premature application of optimization is evil! Only do so when you can't get a better algorithm or CPU, and a profiler reveales that this area is a problem. Goto will only at best save a few cycles from your loop, so it rarely is enough of a solution to help speed problems. Sometimes embedded systems run into those cases where it matters.
|
|
|
|
|
Alex Ngai wrote:
and it will easily destory a program's structure,
and
Alex Ngai wrote:
a heavily-nested loop
So you want to protect the structure of your heavily nested loop code? Isn't that a bit of an oxymoron?
If you are actually concerned with program structure you might consider refactoring the code into a modern object oriented design that utilizes known patterns and techniques to solve simple structure problems like nested loops and 500 line if/elseif/switch statements.
But, that's just my opinion... I could be wrong.
Hate is not a family value
-pete
|
|
|
|
|
Thanks for all you guys's replies,
Unfortunately, at least 3-nested loop is a common practice:
while( )
{
for( ; ; ){
for( ; ; ){
}
}
}
Java does not have goto statement at all, but in order to jump out of a several-nested loop, Java has a smart extended break statement, it works like this:
first:{
second:{
third:{
System.out.println("Before break.");
if(condition)
break second;
System.out.println("This won't execute.");
}
}
System.out.println("after second block.");
}
As we see, it does the job of breaking out of a loop, and it won't have the problem of goto statment which will easily destroy a program's structure. Jave really is a great language, if only it was a system language like C/C++.
-----------------------------
C++ without virtual functions is not OO. Programming with classes but without dynamic binding is called "object based", but not "object oriented".
|
|
|
|
|
Alex Ngai wrote:
Unfortunately, at least 3-nested loop is a common practice:
Smoking crack is a common practice, but what does that have to do with Object Oriented Design? 
"No matter where you go, there your are." - Buckaroo Banzai
-pete
|
|
|
|
|
Seems to me, Java's "smart extended break statement" is nothing but a goto in disguise!
A disguised goto is worse than if it was naked!
Bikram Singh
|
|
|
|
|
At first: my expirience is, if it is going to get too complex, there ist some going wrong. Try to make subfunctions.
A good way to leave loops is to leave the whole funktion with "return" + exitcode.
Another option ist to use a central boolean which in every loop is checked.
Try this @ home. (B&B)
|
|
|
|
|
Well, the return statement only return back to the function calling point, that's breaking out of a function, not breaking out of a nested-loop.
The second option seems interesting, would you explain it a little more? I'm interested.
Thanks
-----------------------------
C++ without virtual functions is not OO. Programming with classes but without dynamic binding is called "object based", but not "object oriented".
|
|
|
|
|
Ravi Bhavnani wrote:
If you can't get yourself to use a goto ... you could instead throw an exception
I disagree. If goto is evil then throwing an exception just to avoid using a goto is even worse! Which one is clearer to the person maintaining the code? Which one are you more likely to get wrong? Sometimes a goto is the right way to do something so long as it's well commented and isn't abused.
|
|
|
|
|
Hi,
I am writing a projet in C++ according to "CCSDS RECOMMENDEDATION FOR LOSSLESS DATA COMPRESSION" that is base on Rice coding algorithm. I am wondering if anybody has the source code that I can referce it.
Thanks a lot!
|
|
|
|
|
|
Hi, Ravi Bhavani,
Thank you for your information. I found this algorithm is different. What I am going to code is bassed on the NASA's Lossless Data Compression recommendation that modified the original Rice
code algorithm and seperated the algorithm into two parts: preprocessor and Adaptive Entropy Coder( http://www.ccsds.org//ccsds/). I searched throught the internet and found the corresponding source code named "SZIP", but I can not get any idea from there. I am wondering if you have heared
anything about this special Rice coding?
|
|
|
|
|
huffmancoding wrote:
I am wondering if you have heared
anything about this special Rice coding?
No, I haven't - sorry.
/ravi
My new year's resolution: 2048 x 1536
Home | Articles | Freeware | Music
ravib@ravib.com
|
|
|
|
|
I placed a 'Reset' button on my property sheet right by the 'OK', 'Cancel', and 'Apply' buttons. I was wondering if anybody knew if there was a standard ID number for a 'Reset' button?
OK - IDOK - 1
Cancel - IDCANCEL - 2
Apply - ID_APPLY_NOW - 12321
Reset - ID_RESET??? - ???
I have currently defined a value of ID_RESET to 10055.
"You're obviously a superstar." - Christian Graus about me - 12 Feb '03
"Obviously ??? You're definitely a superstar!!!" mYkel - 21 Jun '04
Within you lies the power for good - Use it!
|
|
|
|
|
I don't think so. The only "standard" ids I found in winuser.h were:
#define IDOK 1
#define IDCANCEL 2
#define IDABORT 3
#define IDRETRY 4
#define IDIGNORE 5
#define IDYES 6
#define IDNO 7
#if(WINVER >= 0x0400)
#define IDCLOSE 8
#define IDHELP 9
#endif /* WINVER >= 0x0400 */ /ravi
My new year's resolution: 2048 x 1536
Home | Articles | Freeware | Music
ravib@ravib.com
|
|
|
|
|
Thanks Ravi.
I also found those IDs. It is just that ID_APPLY_NOW is not in that list. It is in AfxRes.h along with the other property sheet buttons. So I was just wondering if the reset button was defined some where. I prefer to use standard IDs if there are any.
I will will leave it as it is.
"You're obviously a superstar." - Christian Graus about me - 12 Feb '03
"Obviously ??? You're definitely a superstar!!!" mYkel - 21 Jun '04
Within you lies the power for good - Use it!
|
|
|
|
|
Hello there
As a beginner in windows application programming I've run into a few problems - the most annoying being at a message posting procedure where the message never receives the intended recipient.
I've got a DLL which is intended for posting messages to an dialog at certain events. This posting are carried out by PostMessage. But the problem is that the message never arrives at the dialog. I've used Spy++ to check the handle of the intended recipient and the handle stored in the DLL, and they match. The return code of the PostMessage, is 0 = successful posting, but the dialog never receives the message (I've checked this with Spy++).
If I use HWND_BROADCAST as the first argument for PostMessage, my dialog receives the message correctly, thus i assume that my event handler in the dialog is correct.
What can be wrong?
Best regards
Thomas René Sidor
|
|
|
|
|
If the function fails, the return value is zero. To get extended error information, call GetLastError.
"No matter where you go, there your are." - Buckaroo Banzai
-pete
|
|
|
|
|
I have used GetLastError, and the function itself does not fail. The problem is just that the message never arrives at the intended window.
Best regards
Thomas René Sidor
|
|
|
|
|
PostMessage's return type is BOOL. 0 means FALSE, which means that the message was NOT posted. Again, use GetLastError to work out why.
Stability. What an interesting concept. -- Chris Maunder
|
|
|
|
|
The method does not return 0. Therefore the use of GetLastError just tells that the command was succesfull.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
