From: Scott A C. <sc...@cs...> - 2003-06-27 00:01:08
|
Hello. I recently found a link to this project and I like the idea of it. I hate reimplementing datastructures millions of times, so I'm happy to see the existance of this project. I've got a couple of questions and comments though.. First, is this forked library a derivative of CDK, so it includes all of the enhancements of CDK STDLIB derivative? Second, the few comments. IMHO, its better to have 3 very complete datastructure modules than have 5 half-complete ones. As a user, I'd rather have 3 modules I can use as-id, and implement two of my own than have to enhance 5 modules. Also, may I suggest that we release all of the stable modules and increment the patchlevel as soon as a new module is considered stable and complete. (And we only release (Put all Dev modules into a Stdlib.Dev submodule) Anyways, I have an modified version of the current STDLIB Set module. that supports Set.S.cardinal in constant time, I can improve it so that you can index into a set to get the *n*th element cheaply, and given an element can determine its index from the beginning. (This is useful if you want to use a Set to link to another datastructure.) Interested? If so, how do I get access to CVS? I've never used sourceforge. Scott |