[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: data type polymorphism
From: |
Ivan Vučica |
Subject: |
Re: data type polymorphism |
Date: |
Tue, 24 Mar 2015 22:32:26 +0000 |
If you have a couple hundred lines of code and want to stay in the world of C, you could consider splitting these lines into a .h file, and then #define MYTYPE float, include header, #undef MYTYPE, #define MYTYPE char, include header, etc.
If you are not doing something too computationally intensive, you would use Objective-C classes to wrap your data types. But you are, and so you're dropping to C level to do computation, hence nicest thing to do is to do relatively clean preprocessor tricks.
If you are willing to dilute your code with C++, I like David's approach. :-)
On Tue, Mar 24, 2015, 21:49 Riccardo Mottola <
riccardo.mottola@libero.it> wrote:
Hi Scott,
Scott Christley wrote:
> Hi Riccardo,
>
> Maxthon is not suggesting to pick a type during compile time, but instead use a macro to generate both cases in your code. For your app it would be something like this where the macro has the code, and the C preprocessor is used to insert the code. In reality you still have duplicate code. I was aware of this solution but was interested to see if there was any alternatives.
>
>
> #define PROCESS_IMAGE(DATATYPE) \
> DATATYPE *srcData; \
> DATATYPE *destData; \
> DATATYPE *p1; \
> DATATYPE *p2; \
> // etc.
>
>
> if ([srcImageRep bitsPerSample] == 8)
> {
> PROCESS_IMAGE(unsigned char)
> }
> else if ([srcImageRep bitsPerSample] == 16)
> {
> PROCESS_IMAGE(unsigned short)
> }
>
>
Oh thank you for explaining it. This is feasible for short pieces of
code, but if you have a hundred of lines, either you split it up in
smaller makro-pieces or it is a mess.
Thanks anyway.
Riccardo
_______________________________________________
Discuss-gnustep mailing list
Discuss-gnustep@gnu.org
https://lists.gnu.org/mailman/listinfo/discuss-gnustep
Re: data type polymorphism, David Chisnall, 2015/03/21