RE: [Dev-C++] Help with functions

[Ioannis V. <no...@ya...>]

> -----Original Message-----
> From: dev...@li...=20
> [mailto:dev...@li...] On Behalf=20
> Of Jason Hardman
> Sent: Friday, October 05, 2001 3:32 AM
> To: Dev-C++
> Subject: Re: [Dev-C++] Help with functions
>=20
>=20
> Thanks for pointing out the casting. The only reason I did it=20
> was because I had had to look up the functions and had=20
> noticed the type differences, and I figured it's best to cast=20
> rather than use implicit casting. As for using C style casts=20
> instead of C++, just call me a lazy typer. Using=20
> static_cast<unsigned> is twice as much to type. :)

Yes but the C style casting is equivallent with reinterpret_cast the
most unsafe casting. C++ style casts are as restrictive as possible for
a particular situation and must be used instead of the C cast. It is
ugly in writing because you *very rarely* have to use casts, and casts
may hide potential errors. Much usage of casts in C++ is an indication
of bad code. In a few words, you should almost never need casting in
normal C++ applications.

>=20
> As for sizeof(), it was just the first thing that came to my=20
> head at the time, and as you said it was a const string . As=20
> for using the a C string as opposed to a C++ string=20
> container, that was deliberate. For the task I was performing=20
> I decided that a C string was more efficient than the=20
> alternative, and converting from a string container isn't=20
> hard. I tend to write my C++ programs as more of a hybrid=20
> C/C++ style, as I did learn C first then expanded from it to C++.

The notion that the C subset is fastest than C++ is a myth. See
http://www.research.att.com/~bs/esc99.html and
http://www.research.att.com/~bs/crc.pdf

C++ string is a typedef of basic_string<char> type that is:

typedef basic_string<char> string;

basic_string is a template class and template classes tend to expanded
inline when compiled. Also a good compiler creates instances only of the
member functions of a template class which are used and ignores the
rest. Thus since the code is expanded inline in the application you have
0 runtime overhead.


>=20
> The reason that I had written the sort algorithm was because=20
> someone had asked how to do it. As is my habit I went=20
> overkill on it. I really like taking on problems like that,=20
> although I probably should have let the person figure it out=20
> for himself. Looking at my sortv2 algorithm I think I may=20
> have given him a headache. :-D While I'm no expert on the=20
> standard algorithm's sorting fuctions, I think they mainly=20
> cover sorting into literal order, and not sorting same=20
> letters in to the order they appear in sequence, grouping=20
> them together (skipping spaces), i.e. MISSISSIPPI MUD to=20
> MMIIIISSSSPPUD.

I am currently studying C++ myself too and i haven;t reached at the
algorithms chapters but as far as i know C++ sorting algorithms can be
provided with the rules of how to sort (here the number of appearances
of the letters). Also i think they are template functions too so they
are expanded inline as i mentioned above.


>=20
> Jason.


Keep coding :)
Ioannis

* Ioannis Vranos
* Programming pages: http://www.noicys.f2s.com
* Alternative URL: http://run.to/noicys