Le 17/11/12 18:44, Francisco José Tapia a écrit :
About the compatibility with the boost Statistical Accumulators Library, I didn't thought about it, but it would be useful. I will try to implement, but I need to examine in deep for to provide you a well founded opinion.
About the Policy-Based Data Structures of GCC, I didn't know, but I had learned many things reading the code and the documentation of GCC, and I hope to learn more with them. I say you the same, I will examine in order to provide you a founded opinion.
About the question of sample code.
If you see the online documentation. ( https://dl.dropbox.com/u/8437476/works/countertree/index.html). The last point is the Download. It's a zip file. If you examine this file, you have a example folder with examples of all the data structures, and a benchmark folder with the benchmarks used for the documentation.
Please don't top post on this ML. I have taken a look at the performances on https://dl.dropbox.com/u/8437476/works/countertree/doc/ordered.html#benchmar.... From the figures it seems that the counter tree implementation has worst performances that the std:: when no allocator is used. I'm wondering if the figures for the std:: containers are using a specific allocator? If not, it will be a good idea to add them also, so that the comparison are fair. Best, Vicente