[Soc-2007-dev] Friday status reports!

[Levi Schooley]

I was absent the official first week of SoC, so last week was sort
of the first week for me.

What I have accomplished so far:
1. I have checked out, and am "monitoring," the blender trunk, my
soc branch, and the bmesh branch.
2. I have prepared some documentation (mostly focused on
implementation so far) at:
http://wiki.blender.org/index.php/User:ReD_Fox/SummerOfCode2007
3. I have made some "baby steps" into the source code, and was
able to create a hook in the modifier stack for the bevel tool. I
also did an experiment, and attempted to create
derivedmesh-to-editmesh-and-back-again functions so that I could
use the current bevel code to at least show *something* when the
bevel modifier was applied. No luck so far (although I am learning
more than I initially wanted to about blender's memory allocation
and the evils of The EditMesh).
4. Oh, and yes, I got gdb working (!), and had some fun learning
the ropes of a new tool.
5. Geoffrey Bantle (Briggs) shared his face inset modifier code
with me a couple of days ago, but I have yet to play with it much.

What is planned for next week:
1. Continue familiarizing myself with the source. I will probably
focus on the BMesh API and Blender's current tools code.
2. Finish the derivedmesh-to-editmesh-and-back-again experiment
(for fun, but also as an exercise).
3. Doing a bit more work on the wiki page.
4. Sending out a request for comments on some issues I need to
address (some of those are currently listed on the wiki page, I
believe).
5. And, if I find the time, I will hopefully begin hammering out
some of the needed algorithms.

Problems, questions, and other comments.
1. I need an IDE! Well, OK, I got by without one this past week,
but hopefully I will soon have a new computer with a fresh IDE. It
will make source code browsing quite a bit faster.
2. My internet connection stinks. Hopefully this will be resolved
soon, but I am not able to chat or log much over IRC. If there's
anything important that I need to know at the Sunday meeting (or
at any other time on IRC), I would appreciate it if some kind soul
would pass it on to me via other methods.
3. Regarding design, I have a pretty good idea of what is needed
implementation-wise. I have some rough ideas for the algorithms
that will get me there, but they need to be fleshed out and
refined. For the code design, I still need to put together some
diagrams and information on the wiki, but the basics will be like
the following.
 a) The code will be written using the
currently-in-development-and-integration BMesh API.
 b) Initially, to maintain usability, the tools will be created as
modifiers utilizing conversion functions to and from the
DerivedMesh and BMesh mesh structures. (This may not be the way it
is ultimately implemented, depending on the effects the BMesh
refactor has on DerivedMesh.)
 c) Once the core functions for the bevel modifier are in place,
focus will be switched to utilizing them in an editmode tool. The
mechanism for accomplishing this is still under consideration (it
will likely be direct function calls, since plans are well under
way to replace EditMesh with BMesh).
5. A change to scons on May 31 broke my fragile build environment.
I've yet to resolve the issue (and, thus, have been holding off on
a merge).

I think that's about it!

Levi