Re: [boost] Any interest in a sorting library with a multitude of sorts?

Ok. I will make sure that it is possible for the user to specify the function used to compare objects. However, using the interface class with one's own classes, less-than functions can be defined without interfering with the actual less-than operator. Also, a few of you asked about a "sort within sort". That can already be done by radixsort for your objects because radixsort examines individual bytes, not the whole thing. I knew that people would want this, so I supplied radix sort. I also wrote a function that will put all of the bytes of information in a single sequence, basically, a string of bytes. That way, you can "sort within sort". Also, someone asked if quicksort and radixsort were sub-cases of radix quicksort. Actually, radix quicksort is a merge of the two algorithms. _________________________________________________________________ Play Flexicon: the crossword game that feeds your brain. PLAY now for FREE. http://zone.msn.com/en/flexicon/default.htm?icid=flexicon_hmtagline

On 3/13/07, Sam Schetterer <samthecppman@hotmail.com> wrote:
Ok. I will make sure that it is possible for the user to specify the function used to compare objects. However, using the interface class with one's own classes, less-than functions can be defined without interfering with the actual less-than operator. Also, a few of you asked about a "sort within sort". That can already be done by radixsort for your objects because radixsort examines individual bytes, not the whole thing. I knew that people would want this, so I supplied radix sort. I also wrote a function that will put all of the bytes of information in a single sequence, basically, a string of bytes. That way, you can "sort within sort". Also, someone asked
Isn't that a bit expensive? Why not first compare the first byte/part and then if necessary the second byte/part?
participants (2)
-
Olaf van der Spek
-
Sam Schetterer