Re: [boost] [String algorithm] is_any_of has inefficient implementation

From: Phil Endecott (spam_from_boost_dev_at_[hidden])
Date: 2008-02-15 12:54:15

Giovanni Piero Deretta wrote:
> On Thu, Feb 14, 2008 at 6:50 PM, Phil Endecott
> <spam_from_boost_dev_at_[hidden]> wrote:
>>  In each case, I've used the predicate with std::find_if.  But if I
>>  instead use C's strcspn() to do the whole job, I see a relative
>>  performance for this example of 20.  I find it depressing that the C++
>>  code doesn't come close to that.
>
> Well, it is not really a fair comparison, the 'C' function is likely
> implemented in hand optimized assembler, (http://tinyurl.com/yv8s7g)
> and for such a small function, humans are still better than compilers.
>  So it isn't really C++ versus C. Also the C version is not generic at
> all.

Hi Giovanni,

I can't agree with that.  There's no reason why the C++ implementation 
couldn't be written in hand-optimised assembler too.  Or it could 
simply call the fast C function:

template <> const char* std::find(const char* first, const char* last, 
char value) <const char*> {
   return ::memchr(first,value,last-first);
}

Specialisation allows C++ libraries to provide functions that are both 
generic and optimised for special cases.

Anyway, rather than complaining I should be writing code.  This week I 
have been writing some UTF-8 encoding and decoding and 
Unicode<->iso8859 conversion algorithms.  They seem to be faster than 
the libc implementations which is satisfying especially as I haven't 
even started on the serious optimisations yet.  This will be part of 
the strings-tagged-with-character-sets stuff that I have described 
before.  Anyone interested?

Phil.