sptk



[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: RFCs for next SPTK release



Alexey Parshin wrote:
> 1. What for? To make the code dirty? I really hate it. Most of the
> time, SPTK is used with stl ns, so it would be either sptk:: all over
> the code, or stl:: ..
>  In any case - show me the gains..
Namespaces are for avoiding name collisions.
> 2.
>  a. We need two methods then, const and not const. Will add as soon
> as CVS is working.
Yes, that is correct.
>  b. Making it char requires a conversion in too many places while
> achieves nothing (IMHO)
try compiling following with -pedantic:
long* lp=static_cast<long*>(mybuf.data())
>
> 2006/5/12, Ilya A. Volynets-Evenbakh <ilya@total-knowledge.com>:
>> Hey,
>> Here are few things that would be nice to have.
>>
>> 1. Namespace. Let's wrap all classes in sptk:: namespace
>> 2. Some minor cleanups:
>>     a. make CBuffer.data() non-const (I think it's more or less illegal
>>        to return non-const pointer to member, from const function)
>>     b. Make it return void, not char*, since it is data buffer, not
>>         character buffer.
>>
>> I'll continue this list, as things come to mind.
>>
>> -- 
>> Ilya A. Volynets-Evenbakh
>> Total Knowledge. CTO
>> http://www.total-knowledge.com
>>
>>
>
>

-- 
Ilya A. Volynets-Evenbakh
Total Knowledge. CTO
http://www.total-knowledge.com


List hosted by Total Knowledge

Authoright © Total Knowledge: 2005