This is a version of the red black tree delete file which is based on the third edition of "Introduction to algorithms" by Cormen et al.I compliled BRL-CAD from source and tested.
the rb_delete code patch
raytrace.patch increments the maximum number of solids (ID_MAX_SOLID and ID_MAXIMUM) to 44 and defines the heart constant (ID_HRT) to be 43.
This patch is a modification of the include/raytrace.h file.
db5.patch simply defines a constant DB5_MINORTYPE_BRLCAD_HRT for routine implementing of the v5 database for the heart primitive.
This is a modification of the include/db5.h file.
Issac, you need to provide some instruction here. What am I supposed to do with all of these files? If they're all related, they should be in just one patch file. If they're not related, in what order are they to be applied? Are the patch files instead of the patch.c attachment or in addition to it? Having me guessing or spending time to figure it out is a waste of time.
Also, what does this patch do and why? Saying it's a version from that book doesn't tell me anything. Did our version have a problem? What was it? How did you confirm it? Can you demonstrate the bug and that this actually fixes it?
You need to make this easier to review...
Sean I have corrected db5.patch and raytrace.patch.I made sure they meet the guidelines you indicated in your messages on the mailing list.
As for the rb_delete.patch, I chose to work on this patch when I read the TODO file and it indicated that the rb_delete.c file be upgraded (libbu implementation of rb_delete.c) based on algorithms in the latest version of the Introduction to algorithms book. I have compiled and debugged this patch but I don't exactly know how to demonstrate the bug , verify if the problem existed and so on. At the time, I hadn't learned how to use diff and patch - so that's why its in patch.c format.
Thanks Issac, but I think you might still be missing the point. I had absolutely no question about why you chose to work on this patch. My comment was not an evaluation of your understanding of patch files or related to your GSoC application.
You provided a patch submission to an open source project. Whenever you do that (for ANY project), you need to make it very clear what you've done, why, and submit it in a simple-to-review form. This patch submission is rather messy in that regard and you've still not made it clear with your comment what to do about THIS submission.
The TODO task was specifically to figure out IF we have a problem, to THEN fix the problem. If you did not ascertain whether we have a problem nor whether your patch fixes the problem, what's the point then?? What am I supposed to do with the patch?
Thanks Sean , What about db5.patch and raytrace.patch. Are they correct and/or satisfy completeness.
It is wrong for those patch files to be attached to this patch submission. They have absolutely nothing to do with rb_delete and talking about them here is an inefficient use of time. This is another communication failure. Instead of responding to my questions, you're asking me a question about those completely unrelated files further exacerbating the problem. No, they are not complete -- those simple line changes would likely leave our system in a broken state.
You should not post comments or attachments that have nothing to do with the patch submission. Please answer the questions.
Issac, are you still working on the red-black tree patch? If not, this patch should be closed.
I am no longer working on the red-black tree patch so it can be closed.