> On Sun, 24 Jun 2012, Rodolfo García Peñas wrote:
>> Therefore, not all the .c files have .h file. This is possible, but he
>> main reason is because some of them are using "funcs.h" and
>> "Windowmaker.h" to include their structs and prototypes. Of course,
>> funcs.c and Windowmaker.c doesn't exists :-)
>>
>> For this reason, I think we need "a plan":
>
> You may want to consider that .h files are not necessarily belong to .c
> files and it does not have to be a one-to-one mapping. In the object

Yes yes :-) I know. See the "This is possible..."

> oriented paradigm .h would declare the methods or the interface of an
> object and the .c would contain the implementation (the actual method
> definitions). Thus it's fine to have internal functions defined in the .c
> file and also fine to split the implementation into multiple .c files
> while there's still only one .h file. You could also think of this as
> modules. The functions declared in WindowMaker.h might belong to a core
> module which is implemeted in multiple .c files. Of course this can be
> changed and fixed if this is not anymore the case but the point is that it
> should follow some logic and not just make every .c file have a .h file
> declaring every funtion. That's why we need a plan and have to think about
> what are the modules that need a .h file first :-)

I agree.

Thanks,
kix

> Regards,
> BALATON Zoltan


-- 
Rodolfo García AKA kix
http://www.kix.es/
EA4ERH (@IN80ER)


--
To unsubscribe, send mail to [email protected].

Reply via email to