IRC log for #brlcad on 20120524

00:27.06brlcadstarseeker: so curious about that SIZEOF_*_T suffix.. That symbol from autoconf land is SIZEOF_type where type is the uppercase form of the C type.  The _T isn't really applicable to all sizeof() checks
00:27.11brlcadjust happens to be the suffix for size_t and a few others we check for.
00:27.35brlcadI very well might have wanted to check sizeof(FILE) for example or some X type or something else in order to behave accordingly
00:28.35brlcadactually, one of the checks is supposed to be checking the sizeof(void*), which should be SIZEOF_VOIDP
00:34.16brlcad(and I do think we use that in one or more places)
01:02.06*** join/#brlcad jbschw (~jbschw@ool-4355ee10.dyn.optonline.net)
02:33.52*** join/#brlcad IriX64 (~kvirc@65.95.123.211)
03:15.35CIA-65BRL-CAD: 03phoenixyjll * r50649 10/brlcad/trunk/src/librt/primitives/extrude/extrude_brep.cpp:
03:15.36CIA-65BRL-CAD: Changed the position of the sketch in extrude according to the parameters.
03:15.36CIA-65BRL-CAD: Some comment is added to make it clear.
03:37.09*** join/#brlcad archivist (~archivist@host81-149-189-98.in-addr.btopenworld.com)
03:39.31CIA-65BRL-CAD: 03phoenixyjll * r50650 10/brlcad/trunk/src/librt/primitives/ (revolve/revolve_brep.cpp sketch/sketch_brep.cpp): The old version creates a circle just on the xy plane. Now it's created on the sketch plane.
03:43.51CIA-65BRL-CAD: 03Phoenix 07http://brlcad.org * r3727 10/wiki/User:Phoenix/GSoc2012/Reports: /* Week 1 */
04:39.53*** join/#brlcad anuragmurty (~anurag@14.139.128.11)
06:35.50*** join/#brlcad ksuzee (~ksuzee91@46.149.82.166)
07:20.41anuragmurtybrlcad: hi
08:03.15*** join/#brlcad d_rossberg (~rossberg@BZ.BZFLAG.BZ)
08:22.28anuragmurtyd_rossberg:hi
08:23.14anuragmurtyi wanted to enquire about voxelize
08:24.18*** join/#brlcad stas (~stas@82.208.133.12)
08:55.08d_rossberganuragmurty: morning
08:55.20anuragmurtymorning
08:55.42anuragmurtyare you free right now?
09:10.06d_rossbergno, but don't hesitate to ask though ;)
09:13.46anuragmurtyoh ok.. so i have been through two examples pointed out by brlcad(sean)... one is rtexample, the other is viewweight.. that does give me some hints on how to go about determining what functions will be useful for voxelization
09:15.20*** join/#brlcad kane_ (~kane@e182053064.adsl.alicedsl.de)
09:16.49anuragmurtythere will be rt_hit, rt_miss that i have to modify according to needs.. but what parameters will i need for voxelization? an obvious one would be size of voxel/number of voxels  in each dimension.. i just want to know if i am thinking along the right directions.. obviously things should get clearer after i go through g_qa which samples 3D geometry
09:25.33d_rossbergother parameters could be: rays per voxel, the percentage a voxel has to be filled to take it into account
09:28.59d_rossbergthe algorithm of voxelization probanly involves calling rt_shootray from a grid of start points with grid size = voxel size
09:29.28anuragmurtyok.. i get it i think
09:31.13anuragmurtyand then we can use the ray partitions of rays from different views to decide which voxels to take into account.. ?
09:33.06anuragmurtyi mean using that to calculate volume of that rectangle filled and then deciding..
10:33.53d_rossbergyou don't need different views: e.g. voxel size 1, you hit a component at 17.2 and leave it at 17.8 => 60% of this voxel is filled
10:34.54d_rossbergto improve this you need more rays through this voxel
10:40.06*** join/#brlcad kane_ (~kane@e182053064.adsl.alicedsl.de)
11:16.02*** join/#brlcad kane_ (~kane@e182051111.adsl.alicedsl.de)
11:30.35starseekerhmm, cool:  http://news.sciencemag.org/sciencenow/2012/05/return-of-the-vacuum-tube.html
11:58.27*** join/#brlcad kane_ (~kane@e182051111.adsl.alicedsl.de)
12:05.07*** join/#brlcad kane_ (~kane@e182051111.adsl.alicedsl.de)
12:10.52*** join/#brlcad kane_ (~kane@e182051111.adsl.alicedsl.de)
12:12.42CIA-65BRL-CAD: 03d_rossberg * r50651 10/brlcad/trunk/src/conv/ (10 files in 2 dirs): a raw (simple ASCII bag of triangles) to g geometry converter
13:22.32CIA-65BRL-CAD: 03d_rossberg * r50652 10/brlcad/trunk/ (4 files in 2 dirs):
13:22.32CIA-65BRL-CAD: apply sourceforge parch 3527658 (libbn_refactor_tolerance.diff) from kane_:
13:22.32CIA-65BRL-CAD: "refactor and manage libbn tolerance uses by providing an interface default
13:22.32CIA-65BRL-CAD: (e.g., an init macro) and making everyone use that where it is hardcoded to
13:22.32CIA-65BRL-CAD: 0.0005 presently"
13:39.04d_rossbergkane_: i've just realized that there is a BN_TOL_INIT() macro which could be used to initialize bn_tol with default values
13:40.11d_rossbergto initialize the struct to zero there is another BN_TOL_INIT_ZERO() macro
13:49.48CIA-65BRL-CAD: 03Anuragmurty 07http://brlcad.org * r3728 10/wiki/User:Anuragmurty: /* community bonding period */
13:50.21*** join/#brlcad anuragmurty (~anurag@14.139.128.12)
14:40.31*** join/#brlcad kane__ (~kane@e182051111.adsl.alicedsl.de)
15:19.12brlcadanuragmurty: thanks and welcome (again)
15:20.50*** join/#brlcad Stattrav_ (~Stattrav@61.12.114.82)
15:26.20CIA-65BRL-CAD: 03starseeker * r50653 10/brlcad/trunk/ (CMakeLists.txt misc/CMake/BRLCAD_CheckFunctions.cmake): This should improve the behavior and robustness of the type checking macro in BRL-CAD. While we're at it, make the variable names automatic and deduced from the type itself.
15:36.40*** part/#brlcad kane_ (~kane@e182051111.adsl.alicedsl.de)
15:44.36brlcadanuragmurty: do you have a good idea what your next steps need to be?
15:45.05brlcadrather, a specific action to be taken next
15:52.17*** join/#brlcad jbschw (~jbschw@ool-4355ee10.dyn.optonline.net)
16:23.28CIA-65BRL-CAD: 03n_reed * r50654 10/brlcad/trunk/src/other/step/src/cleditor/ (STEPfile.cc STEPfile.h seeinfodefault.h): Apply changes from SCL git 4858af2 removing code supporting step files (N279 files) that pre-date ISO 10303-21:1994.
16:27.10anuragmurtybrl-cad: hi! was gone for dinner
16:27.37brlcadpretty sure that's allowed ;)
16:27.48anuragmurtyi was able to talk to daniel today
16:28.10anuragmurtyrt_shootray for a grid
16:28.13brlcadirc isn't always instantaneous, meant to lurk just like you did, see messages when you get back, sometimes interactive othertimes not
16:28.18brlcadyeah, I read
16:28.39anuragmurtyyes.. so thats what i plan to do..
16:28.51anuragmurtywill contact if i hit a roadblock
16:29.18anuragmurtyright now i don't think a very basic program would differ much from rtexample
16:29.32anuragmurtymaybe i am wrong.. but thats what i am tryin
16:30.04brlcadprobably won't but that's exactly the sort of progress you should post daily to the patches tracker until you have commit
16:30.09brlcadand document in your dev log on the wiki
16:30.16brlcadeven if it's just one sentence
16:30.41brlcadI'd suggest starting with something very basic, like firing one ray ala rtexample
16:31.39anuragmurtyhmm.. got it.. will do that right now..
16:31.41brlcador maybe fire a ray down three axes
16:31.51brlcadthat could be the second step
16:32.16brlcadthat would effectively triple sample a single cell
16:32.58anuragmurtyyes.. i will do that.. i had another doubt..
16:33.19anuragmurtyfiring from 3 different views v. 3 from the same direction
16:33.31anuragmurtysame view i mean.. which is better??
16:33.39brlcadunderstand daniel's point, though -- you certainly don't need more than one
16:34.13brlcad3 from the same direction means what?
16:34.18anuragmurtysame view..
16:34.26brlcad3 from the same view means what?
16:34.33brlcad3 what?
16:34.36anuragmurtyrays
16:34.43anuragmurtyi think i am messing something up
16:34.45brlcadwhy would you shoot 3 rays?
16:34.54anuragmurtyyes i now realise
16:34.57anuragmurtysorry abt that
16:35.07brlcadno need to apologize
16:35.35anuragmurtyok.. thanks though.. dunno  what i was thinking
16:35.38anuragmurty:-/
16:35.51brlcadyou might want to shoot multiple rays to sample a voxel cell, that would be why you might shoot N rays
16:36.04brlcadbut certainly not the same ray, or you'll just get the same information three times
16:36.27anuragmurtyexactly.. i think i meant something like no. of rays per voxel vs. smaller voxels
16:36.28brlcadrays are infinitely thin -- they give you a thickness
16:36.48anuragmurtybut again.. i guess i will think first
16:41.41brlcadjust make sure that thinking also involves discussion :)
16:42.08anuragmurtyok.. :)
16:42.15brlcadthis isn't like doing homework -- you're expected to interact as we have a vested interest in what you're doing
16:42.33brlcadyou're not getting graded, you're making something others will use
16:42.48brlcadso at a minimum, you need to discuss it because we need to understand what you're doing too
16:43.41anuragmurtyi will ensure i do that..
16:45.01brlcadso I think a good goal to have for a few weeks out is getting whatever technique working for shooting rays, and then getting that into a command in mged/archer
16:45.39brlcadfrom the discussion on the mailing list a couple days ago, it'll be good to include conversion to voxel form as part of our conversion.sh script so we can validate conversions
16:46.07brlcadwhich means creating a 'vol' object for a given input geometry
16:47.23brlcadjust something to keep in mind, since the first steps are more oriented around learning the rt_shootray() interface
16:48.18brlcadyou need to be able to shoot a grid at a minimum, so perhaps that's best to do first
16:50.08anuragmurtyhmm.. i think i am sort of comfortable with the shootray function.. will try out the 3 different views first, then the grid should follow..
16:50.36anuragmurtywill look at the conversions after i am done with these
16:51.19brlcaddefinitely post your sources now and then after you shoot three views (then after a grid) .. those are good checkpoints
16:51.58anuragmurtyok.. it will be on my log..
16:52.13brlcadhm?
16:52.31anuragmurtyoh sources.. yes
16:52.59anuragmurtyi will do that..
16:53.02brlcadgreat
16:53.19brlcadboth should be every day, doesn't even matter if you're done -- it's about communicating your effort and progress
16:53.36brlcadboth being log and patch/commit
16:54.05anuragmurtyok..
16:54.15brlcadif you have any feedback on patches already in the tracker, those should take priority because we want to get you reviewed and access granted
16:54.18brlcadso you can be more efficient
16:54.37brlcadit's also a lot easier to review commits than it is patches ;)
16:54.59anuragmurty:)
16:55.58brlcadso yeah, post very very small incremental steps as patches and respond to any feedback with updates -- getting commit set up is the #1 priority
16:57.16anuragmurtyi will do that.. actually d_rossberg is reviewing the patch i sent.. some problems, but should be done soon..
17:01.10brlcadexcellent, I meant more in general
17:01.22brlcadand not just for you, that's all students that don't yet have commit access
17:02.22brlcadthis is the first time several students didn't put in effort during bonding period to obtain commit, so we need to get that sorted out quickly
17:02.43brlcadshould have been done a month ago.. :)
17:25.51*** part/#brlcad anuragmurty (~anurag@14.139.128.12)
17:33.42*** join/#brlcad Stattrav_ (~Stattrav@61.12.114.82)
18:24.14CIA-65BRL-CAD: 03Crdueck 07http://brlcad.org * r3729 10/wiki/User:Crdueck/log:
19:00.45CIA-65BRL-CAD: 03tbrowder2 * r50655 10/brlcad/trunk/src/libbu/test_vls.c: add new tests, some format tidying
19:10.46*** join/#brlcad cristina (~quassel@188.24.79.30)
19:11.52*** join/#brlcad cristina (~quassel@unaffiliated/cristina)
19:16.34*** join/#brlcad cristina (~quassel@188.24.79.30)
19:16.51*** join/#brlcad cristina (~quassel@unaffiliated/cristina)
19:22.48CIA-65BRL-CAD: 03tbrowder2 * r50656 10/brlcad/trunk/src/libbu/vls.c: correct operator for combining flags
19:24.40CIA-65BRL-CAD: 03tbrowder2 * r50657 10/brlcad/trunk/src/libbu/vls.c: reformat for ease of seeing boolean choices
19:26.20CIA-65BRL-CAD: 03tbrowder2 * r50658 10/brlcad/trunk/src/libbu/vls.c: add another legitimate snprintf flag charcacter
19:30.33CIA-65BRL-CAD: 03tbrowder2 * r50659 10/brlcad/trunk/src/libbu/vls.c: eliminate check for field type here (even though it's a hybrid length modifier)
19:34.54CIA-65BRL-CAD: 03tbrowder2 * r50660 10/brlcad/trunk/src/libbu/vls.c: move all length modifier checks to end of format specifier parse, add comment to that effect
19:38.26CIA-65BRL-CAD: 03tbrowder2 * r50661 10/brlcad/trunk/src/libbu/vls.c: correct position of comment
19:48.02CIA-65BRL-CAD: 03tbrowder2 * r50662 10/brlcad/trunk/src/libbu/vls.c: reorder length mod defs, add new length group
19:51.04CIA-65BRL-CAD: 03tbrowder2 * r50663 10/brlcad/trunk/src/libbu/vls.c: remove spurious spaces
19:58.40CIA-65BRL-CAD: 03tbrowder2 * r50664 10/brlcad/trunk/src/libbu/vls.c: don't clear length flags for 'l' or 'h' UNTIL we check for doubling--this should fix the 32-bit failure
20:45.22CIA-65BRL-CAD: 03tbrowder2 * r50665 10/brlcad/trunk/TODO: 32-bit vls failure fixed with rev 50664, confirmed on 32-bit Linux Mint LMDE
20:48.53CIA-65BRL-CAD: 03tbrowder2 * r50666 10/brlcad/trunk/src/libbu/vls.c: reorder vars for ease of maintenance
20:50.18CIA-65BRL-CAD: 03tbrowder2 * r50667 10/brlcad/trunk/src/libbu/vls.c: clarify comment
20:53.24CIA-65BRL-CAD: 03tbrowder2 * r50668 10/brlcad/trunk/src/libbu/vls.c: align vars, correct spacing, clarify comment
20:56.34CIA-65BRL-CAD: 03tbrowder2 * r50669 10/brlcad/trunk/src/libbu/vls.c: eliminate unnecessary check
20:57.24CIA-65BRL-CAD: 03Stattrav 07http://brlcad.org * r3730 10/wiki/User:Stattrav/GSoC2012_log: Updation of logs for day 24
20:59.15CIA-65BRL-CAD: 03tbrowder2 * r50670 10/brlcad/trunk/src/libbu/vls.c: add another valid char possibility
21:03.58CIA-65BRL-CAD: 03tbrowder2 * r50671 10/brlcad/trunk/src/libbu/vls.c: expand comment for clarity
21:06.28CIA-65BRL-CAD: 03tbrowder2 * r50672 10/brlcad/trunk/src/libbu/vls.c: ws
21:09.52CIA-65BRL-CAD: 03tbrowder2 * r50673 10/brlcad/trunk/src/libbu/vls.c: style, ws
21:10.58CIA-65BRL-CAD: 03tbrowder2 * r50674 10/brlcad/trunk/src/libbu/vls.c: typo in comment expansion
21:18.57CIA-65BRL-CAD: 03tbrowder2 * r50675 10/brlcad/trunk/src/libbu/vls.c: use a var for clarity and ease of maintenance
21:21.46*** join/#brlcad piksi (piksi@pi-xi.net)
21:24.43CIA-65BRL-CAD: 03tbrowder2 * r50676 10/brlcad/trunk/src/libbu/vls.c: more idiomatic
21:26.05CIA-65BRL-CAD: 03tbrowder2 * r50677 10/brlcad/trunk/src/libbu/vls.c: add comment
21:30.34*** join/#brlcad kane__ (~kane@e182051111.adsl.alicedsl.de)
21:31.11CIA-65BRL-CAD: 03tbrowder2 * r50678 10/brlcad/trunk/src/libbu/vls.c: tidy one comment, move another
21:31.31*** part/#brlcad kane__ (~kane@e182051111.adsl.alicedsl.de)
21:36.20CIA-65BRL-CAD: 03tbrowder2 * r50679 10/brlcad/trunk/src/libbu/vls.c: combine decl with initialization
21:40.55CIA-65BRL-CAD: 03tbrowder2 * r50680 10/brlcad/trunk/src/libbu/vls.c: combine decl with initialization
21:43.55CIA-65BRL-CAD: 03tbrowder2 * r50681 10/brlcad/trunk/src/libbu/vls.c: ws
22:05.37*** join/#brlcad Al_Da_Best (~Al_Da_Bes@elvyn-248-109.halls.student.lut.ac.uk)
22:28.47*** join/#brlcad jbschw (~jbschw@ool-4355ee10.dyn.optonline.net)
22:35.00*** join/#brlcad Al_Da_Best (~Al_Da_Bes@elvyn-248-109.halls.student.lut.ac.uk)
22:45.28*** join/#brlcad cristina_ (~quassel@188.24.71.175)
22:57.30CIA-65BRL-CAD: 03n_reed * r50682 10/brlcad/trunk/src/other/step/src/ (15 files in 2 dirs): apply SCL git e72ca29; remove some superfluous boolean types and use std bool more often

Generated by irclog2html.pl Modified by Tim Riker to work with infobot.