IRC log for #brlcad on 20100623

00:06.55*** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1)
00:31.57CIA-40BRL-CAD: 03brlcad * r39670 10/brlcad/trunk/src/proc-db/spiral.pl:
00:31.58CIA-40BRL-CAD: add a fudge factor of 6 degrees to compensate for the fact that we're
00:31.58CIA-40BRL-CAD: translating the segments based on their lower corner point instead of their
00:31.59CIA-40BRL-CAD: natural center. the x/y needs to change and/or the rotation angle needs to take
00:31.59CIA-40BRL-CAD: the size and keypoint of the box into consideration. fudge is 'good enough' for
00:32.00CIA-40BRL-CAD: this scripting example, though.
00:32.00CIA-40BRL-CAD: we also batch together mged invocations in sets of 100 to reduce overhead. improves runtime by two orders of magnitude. names are simplified as well.
00:36.07brlcadpurdy!
00:36.26brlcadlast render for the day under way
00:39.38starseekerkanzure: hah - HeeksCAD does compile with a Makefile tweak
00:39.44starseekernice
00:45.25kanzureneato
00:47.00starseekerbemusedly watches it try to open the OpenMoko step file...
00:47.54kanzureheh
00:49.50starseekercool - justin got his masters degree and is resuming work on gcam
00:57.01Ralithoo
01:05.26*** join/#brlcad yukonbob (~svs@S0106001125477e9c.ok.shawcable.net)
01:09.44starseekerum
01:09.45starseekercan't read "rateknobs": no such variable
01:09.45starseekerMGED unable to initialize gui, reverting to classic mode.
01:11.25starseekerO.O what the...
01:11.29starseekerred works here
01:11.32starseekeraaaaaaaugh
01:12.30starseekeroh, there we go...
01:12.32starseekerkinda works
01:12.34starseekerodd...
01:27.34brlcadhttp://brlcad.org/tmp/spiral.png
01:28.51brlcadand that's enough of that :)
01:29.03yukonbobhello, #brlcad
01:30.06yukonbobq: again  -- (don't recall having seen an answer over last two days): anybody have ideas for converting a CSG miter joint to a 2d plot for application on a real world object?
01:32.16yukonbobeg: imagine two pipes joined nearly perpendicularly, like a "T" (but not necessarily 90 degrees). The "stem" (versus "top crossbar") of the T will need a miter joint. If I model in BRLCAD, ideas for generating a piece of paper I can wrap around ereal-world object to guide miter cut?
02:19.03CIA-40BRL-CAD: 03brlcad * r39671 10/brlcad/trunk/src/proc-db/spiral.pl: cleanup and explanation for example purposes
02:21.45CIA-40BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/upload: uploaded "[[Image:Spiral.png]]": Example procedural geometry model generated by a Perl script, rendered with BRL-CAD.
02:31.19CIA-40BRL-CAD: 03Sean 07http://brlcad.org * r2246 10/wiki/Spiral: initial discussion writeup for the example spiral procedural geometry script
02:32.14brlcadkanzure: fyi, http://brlcad.org/wiki/Spiral
02:33.04brlcadyukonbob: the only ideas I have involve writing code
02:34.20brlcadsounds like a somewhat complex mapping of an arbitrary UV surface space to 2D
02:34.49kanzurethanks
02:35.39brlcadyukonbob: you could script the projection if you know that it's cylindrical, taking the cut geometry and sampling in a cylindrical pattern and then writing the output to a 2D projection as you go
02:38.16yukonbobhi brlcad :)
02:38.39yukonbobya -- I didn't figure it'd be easy -- but it got more complex as I imagined a bit harder...
02:39.06yukonbobhrmm.. I take that back... :)
02:39.38yukonbobI am imagining pipes (at the moment); so at least it's a "regular" object...
02:39.59yukonbobone could then shoot rays "dead on" and make measurements...
02:40.04yukonbobis happy
02:51.55brlcadnirt was sort of designed for that sort of scriptable pattern sampling
03:03.07brlcadenvisioning a script that iterates over azimuth and height, shooting rays and for each ray that hits at a prescribed depth, you write out a pixel and proceed to the next cell
03:05.58CIA-40BRL-CAD: 03brlcad * r39672 10/brlcad/trunk/src/proc-db/spiral.pl: minor explanation consistency
03:54.04Ralithhuh.
03:54.11Raliththere's no BRL-CAD package in Arch.
03:54.18Raliththat's mildly distressing.
04:30.10Ralithconsiders putting together an AUR package
04:38.26Ralithbrlcad: is there a description of the function of all BRL-CAD's (library) components somewhere?
07:36.38Ralithjust tried to build SVN
07:36.46Ralithhm
07:36.47Ralithcleans
08:48.57*** join/#brlcad mafm (~mafm@83.45.73.17)
08:49.43*** join/#brlcad Ralith (~ralith@216.162.199.202)
09:04.15brlcadRalith: there is a description in HACKING, src/README, and (to a lesser extent) doc/PROJECTS
09:04.44Raliththanks
09:07.35mafmbrlcad: didn't you use JAMA in your project, the Java version?
09:08.31brlcadin one place related to nurbs evaluation, not sure it's still being used
09:09.57Ralithwhere would I find detailed API docs for libwdb?
09:16.52mafmwell, I have to not-used in my program, used Apache Commons Math instead
09:17.07mafmit had some weird bugs
09:17.20mafmthe bad part is that even CERN libs copy from it :S
10:34.27*** join/#brlcad Nohla_ (~Nohla@168.226.177.20)
11:05.58d-loMernin all!
11:45.41brlcadRalith: why, from the source of course
11:46.14brlcadactually, there's a manual page (man libwdb)
11:46.21brlcadplus lots of examples in src/proc-db
12:06.23*** join/#brlcad parigaudi (~quassel@pd95b7f5e.dip0.t-ipconnect.de)
12:36.15CIA-40BRL-CAD: 03d_rossberg * r39673 10/brlcad/trunk/src/libged/CMakeLists.txt: synced with Makefile.am by adding combmem.c to the build
12:43.41*** join/#brlcad Stattrav (~Stattrav@117.192.156.213)
14:57.17CIA-40BRL-CAD: 03r_weiss * r39674 10/brlcad/trunk/src/conv/obj-g_new.c: added test_face options, fixed a couple logic bugs, added nmg_fix_normals for closed shells
15:12.25*** join/#brlcad Stattrav (~Stattrav@117.192.149.87)
15:14.30starseekerBah.
15:14.59starseekerthe parsing of the temporary file for red is weak - the addition of other elements in the file messes it up
15:16.40starseekerthinks a little...
15:16.42starseekerhmm...
15:22.56starseekerbrlcad: edcodes seem to apply a change here - how did you get it to fail?
15:23.31starseekeroh, edcomb
15:23.33starseekernevermind
15:23.36starseekercarry on
15:25.15starseekerhmm... edcomb does do something here, although admittedly it's not the most easy to use command I've ever seen
15:47.10*** join/#brlcad Nohla_ (~Nohla@168.226.179.80)
16:25.08*** join/#brlcad Nohla_ (~Nohla@168.226.179.80)
17:01.42``Erikthey used all my dang charcoal O.o
17:18.50``Erikmilton irl: http://idle.slashdot.org/story/10/06/23/151227/Woman-Jailed-For-Starting-Office-Fire-To-Leave-Work-Early?art_pos=4
19:11.50*** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1)
19:39.06CIA-40BRL-CAD: 03erikgreenwald * r39675 10/brlcad/trunk/src/librt/primitives/metaball/metaball.c: steal the UV method and jam it into rt_metaball_uv.
20:34.01RalithI'm getting TCL errors trying to install a freshly built SVN.  I have a system TCL (version 8.5.8) and am getting "/home/ralith/src/brlcad/brlcad/src/other/tcl/library/init.tcl: version conflict for package "Tcl": have 8.5.6, need exactly 8.5.8
20:34.06Ralith"
20:38.19Ralithconfig.log seems to indicate that system TCL tests passed
20:43.48Ralithreran configure, verified it's using system TCL/Tk
20:44.26Raliththis is in tkhtml3
20:47.36Raliththere appears to be cruft lying around the svn dir somewhere
20:48.27Ralithchecks out fresh
20:52.40brlcadstarseeker: I was trying to turn an existing comb into a region, not modify an existing region
20:52.44brlcadthat might have been the flaw
20:53.06starseekerwith edcomb?
20:53.11brlcadyep
20:53.53starseekermy thought on edcomb is to make it take attr=val pairs on the command line to make it more flexible, but I don't know for sure
20:53.55brlcadcuriously don't know of any other way to create a region out of a list of objects
20:54.34brlcadwanted something like "r foo.r *.s" .. but that obviously doesn't work because r expects operators
20:54.35starseekerhmm... - the g command followed by an attr set?
20:54.42brlcadright
20:55.02brlcadwhich is basically what I ended up doing, but really didn't want to set attr directly
20:55.26starseekerhrm
20:55.28brlcadas there are other values assocated with regions like the material_id, los, region_id, etc
20:55.32starseekerwe could make r more flexible...
20:55.43brlcador g
20:56.01brlcadg -r foo.r *.c
20:56.06starseekernods
20:56.32starseeker(or both - probably reasonable for both commands to be more flexible)
20:57.04starseekeris currently setting up some mechanisms to deal with standard attributes more easily before rewiring red.c...
20:57.38brlcadRalith: sounds like tkhtml is looking in our tcl source dir (probably because it loads tcl.m4 or tclConfig.sh from there) instead of the system dir
20:59.01Ralithyeah, closer examination showed that it had some state left over from very old compilations
20:59.05Ralithwhich make distclean failed to wipe
21:03.30CIA-40BRL-CAD: 03brlcad * r39676 10/brlcad/trunk/doc/docbook/lessons/es/ (8 files in 2 dirs): beneficent jesica completes translation of lesson 9 from english to spanish. that just leaves two to go! awesome.
21:05.39CIA-40BRL-CAD: 03brlcad * r39677 10/brlcad/trunk/doc/docbook/lessons/es/ (4 files): remove the executable bit.
21:12.44brlcadit's about time to take the docs to the next level and get new pdfs up on the website
21:13.12Ralithoo
21:14.06CIA-40BRL-CAD: 03brlcad * r39678 10/brlcad/trunk/NEWS: jesica provided spanish translations for lessons 4, 5, 7, 8, 9, 10, 11, 12, 13, and 14 of the mged vol II tutorials! only two remaining
21:30.52starseekerbrlcad: you mean tackle the problem of proper Docbook stylesheets?
21:34.57brlcad4yup
21:36.18starseekerturns slightly pale
21:36.23Ralithheh
21:39.38brlcadmighty whitie!
21:40.18starseekerheh - yeah, I guess getting more pale would be a trick for me
21:40.28starseekeravoids the day star
22:34.28``Erikaday star evil
22:38.16Nohla_brlcad, lied you again :P
22:38.33Nohla_I'll send you the 15th in a few minutes
22:39.46Nohla_fortunately, I can only do this once more
22:43.33*** join/#brlcad Nohla (~Nohla@168.226.179.80)
22:43.48CIA-40BRL-CAD: 03starseeker * r39679 10/brlcad/trunk/ (include/db5.h include/raytrace.h src/librt/db5_types.c):
22:43.48CIA-40BRL-CAD: Alright, this isn't close to ready yet but has definitely reached the 'I don't
22:43.55CIA-40BRL-CAD: want to have to re-create it' stage - commit initial work on some basic routines
22:43.55CIA-40BRL-CAD: for syncing between the comb datastructure and avs attributes, with some helper
22:43.55CIA-40BRL-CAD: functions to handle the various naming convention stuff. These functions will
22:43.55CIA-40BRL-CAD: be used initially to clean up the red.c logic, but will probably be needed in
22:43.56CIA-40BRL-CAD: other places where standard attributes are changed.
22:44.23CIA-40BRL-CAD: 03r_weiss * r39680 10/brlcad/trunk/src/conv/obj-g_new.c: added ability to stop on nmg bomb instead of always falling back to native-bot, added code to partly remove zero length edges
22:46.25starseekershould have moved the sanity checking of attribute values to db5_standardize_avs before committing... whoops. Oh well, after supper
22:47.48starseekerbrlcad: actually, do we already have a packaged routine for parsing out the r/g/b color strings?  Or if not, should that ability be in libbu in case other routines want it?
22:59.09``Erikheh /usr/local/etc/periodic/daily/600.genbrlcaddocs
23:00.36``Erikint color[3]; sscanf(buf,"%i/%i/%i",color+0,color+1,color+2); ?
23:01.15``Erik(or &color[0], &color[1], &color[2] if ptr math scares ya)
23:01.51Ralithseems pretty trivial
23:02.34brlcadNohla: jajajaja
23:02.46brlcad"Leo la palabra truck una vez mas hoy y exploto" jaja, awesome
23:03.03``Erikralith: I've seen that tcl exact issue, removing the -exact in init.tcl seems to work ok
23:03.09Nohlaseee ¬¬
23:04.11Ralith``Erik: already most of the way through a recompile, and there probably would have been other fallout from the stale data anyway
23:04.18``Erikkapow O.o
23:04.19brlcadstarseeker: yes, rt_comb_get_color() as well as routines for reading/writing color tables
23:05.04brlcadwould like to move AWAY from version-specific API names .. "db5"
23:06.09brlcadthe fact that those are stored as attributes should probably be an implementation detail, with added routines for the ones we don't remove like get_color
23:07.04brlcadneed to think about it some more, though
23:29.09*** join/#brlcad louipc (~louipc@archlinux/trusteduser/louipc)
23:33.02*** join/#brlcad akafubu (~akafubu@unaffiliated/akafubu)
23:59.13*** join/#brlcad Nohla (~Nohla@168.226.179.80)

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