IRC log for #brlcad on 20110629

00:09.47brlcadnot surprising
00:09.54CIA-62BRL-CAD: 03brlcad * r45277 10/brlcad/trunk/src/libged/ (100 files): remainder of ws indent consistency cleanup. reorder to eliminate some forward decls too.
00:11.05brlcadimproving code quality is always fair game
00:12.05brlcadwhether it's some bit of logic that could be cleaned up or a usage pattern that begs refactoring or outright crap code that needs to be taken out back and shot
00:14.13CIA-62BRL-CAD: 03brlcad * r45278 10/brlcad/trunk/src/libged/ (CMakeLists.txt Makefile.am subtype.c): subtyping is something that should be defined within the objects themselves down in librt, not at this high level. no matter, the code is a ways off from compiling anyways, so removed from dist.
00:15.00*** join/#brlcad merzo (~merzo@157-86-133-95.pool.ukrtel.net)
00:15.41CIA-62BRL-CAD: 03brlcad * r45279 10/brlcad/trunk/src/libged/ (vdraw.c wdb_vdraw.c): vmath provides M_SQRT2
00:23.33CIA-62BRL-CAD: 03brlcad * r45280 10/brlcad/trunk/src/libged/adc.c: remove ged_ prefix on non-public functions, reorder to eliminate forward decls
00:30.31CIA-62BRL-CAD: 03brlcad * r45281 10/brlcad/trunk/src/libged/ged.c: don't blindly close the wdbp
00:38.49CIA-62BRL-CAD: 03brlcad * r45282 10/brlcad/trunk/src/libged/ged.c: initialize the other struct ged members during ged_init() so they can be properly memory-managed.
00:39.34CIA-62BRL-CAD: 03brlcad * r45283 10/brlcad/trunk/src/libged/ged.c: bah, fix typos
00:46.34*** join/#brlcad crazy_imp (~mj@a89-182-231-95.net-htp.de)
00:50.03bhinesleybrlcad: yeah, I should have taken a closer look at otranslate/ptranslate. That helps a lot.
01:14.24CIA-62BRL-CAD: 03bhinesley * r45284 10/brlcad/trunk/src/libged/ (path.c translate.c): ensure path is initialized and isn't / before trying to access it's directories
01:24.06CIA-62BRL-CAD: 03bhinesley * r45285 10/brlcad/trunk/src/libged/translate.c: translate.c accidentally included in last commit... inverse merge 45284
02:44.30CIA-62BRL-CAD: 03bhinesley * r45286 10/brlcad/trunk/src/librt/db_fullpath.c: it is already assumed that an empty string corresponds to '/', so handle a NULL string the same way (rather than bombing)
02:48.49CIA-62BRL-CAD: 03brlcad * r45287 10/brlcad/trunk/src/librt/wdb.c: if we're closing a wdbp, make sure all memory is released. set the magic to 0 for sanity too in case someone looks at that memory again later.
04:08.47*** join/#brlcad DarkCalf (DC@2002:ade7:2862::ade7:2862)
04:24.56brlcadstarseeker: definitely related, though how to fix this correctly is going to be very tricky .. spagetti code intermingled on the tcl and c sides, both trying to manage that wdbp
04:53.01CIA-62BRL-CAD: 03bhinesley * r45288 10/brlcad/trunk/src/ (libged/translate.c tclscripts/archer/ArcherCore.tcl): tranlation of all objects in a single combination's tree is now working: 'translate -r 5 5 5 / obj1.c', or equivalently 'translate 5 5 5 obj1.c'
05:04.23*** join/#brlcad DarkCalf (DC@173.231.40.98)
05:21.14*** join/#brlcad juanman (~quassel@unaffiliated/juanman)
05:45.00CIA-62BRL-CAD: 0399.125.86.110 07http://brlcad.org * r2939 10/wiki/User:Bhinesley: /* Log */ Friday, yesterday, today, plan for the week
05:54.12bhinesleybrlcad: I'm getting a segfault when I run 'make'. I see that you might still be working on it, so I'll just say that after an inverse merge of r45287 it goes away. http://pastebin.mozilla.org/1260422
06:20.30CIA-62BRL-CAD: 03brlcad * r45289 10/brlcad/trunk/src/librt/wdb.c: better wdb init. don't just set the magic, the forw/back list pointers need initializing. init both vls, not just one of them.
06:24.31CIA-62BRL-CAD: 03brlcad * r45290 10/brlcad/trunk/src/librt/wdb.c: set the magic properly useing BU_LIST_MAGIC_SET()
06:25.50CIA-62BRL-CAD: 03brlcad * r45291 10/brlcad/trunk/src/libged/wdb_obj.c: don't wdb_close does the dequeue and vls frees for us
06:26.56CIA-62BRL-CAD: 03brlcad * r45292 10/brlcad/trunk/src/libged/ged.c: be careful the gdp is not already allocated (should get set to null when deallocated)
06:35.24*** join/#brlcad Stattrav (~Stattrav@122.167.214.98)
06:35.24*** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav)
06:42.27CIA-62BRL-CAD: 03brlcad * r45293 10/brlcad/trunk/src/mged/setup.c: protect this file from auto-formatting by making the expansion of ',' into a compile-time error. the 35,25 and 45,45 commands along with the mged_display tcl variable relies on there being no spaces around the comma.
06:48.05CIA-62BRL-CAD: 03brlcad * r45294 10/brlcad/trunk/src/mged/mged.c: (log message trimmed)
06:48.05CIA-62BRL-CAD: the wdb tcl objects and ged structures both assume they get to own the wdbp they
06:48.05CIA-62BRL-CAD: were passed. can't just remove the wdb_close() from wdb_deleteProc() as there
06:48.05CIA-62BRL-CAD: doesn't seem to be any place in mged where the .inmem wdb is actually recorded
06:48.05CIA-62BRL-CAD: (it's stored inside a tcl callback). so let the tcl objects do their thing, but
06:48.05CIA-62BRL-CAD: then that means the ged needs to acquire a full-fledged copy of the wdb and gets
06:48.06CIA-62BRL-CAD: treated as another dbi observer. this requires more extensive testing for
06:49.26brlcadstarseeker: that should fix it, but it definitely needs more testing (particularly memory testing) ... the open/close code is a royal mess, so it really warrants a valgrinding
06:59.22CIA-62BRL-CAD: 03brlcad * r45295 10/brlcad/trunk/src/libged/analyze.c: reorder in order to eliminate forward references and disambiguate private functions from public API by not using ged_ prefix.
07:06.50CIA-62BRL-CAD: 03brlcad * r45296 10/brlcad/trunk/src/libged/ (dg_obj.c vdraw.c wdb_vdraw.c): reorder and rename in order to remove the ged_ prefix on non-API functions. disambiguate with the tcl-based vdraw_cmd(), renaming to vdraw_cmd_tcl().
07:21.13*** join/#brlcad merzo (~merzo@193.254.217.44)
10:52.19*** join/#brlcad d_rossberg (~rossberg@BZ.BZFLAG.BZ)
11:32.53starseekerbrlcad: so should we hold off on the release for now?
11:33.08*** join/#brlcad kunigami (~kunigami@loco-gw.ic.unicamp.br)
11:36.10*** join/#brlcad piksi (piksi@pi-xi.net)
12:08.53*** join/#brlcad juanman (~quassel@unaffiliated/juanman)
12:28.10*** join/#brlcad dtidrow (~dtidrow@c-68-84-167-135.hsd1.mi.comcast.net)
13:02.30*** join/#brlcad juanman (~quassel@unaffiliated/juanman)
13:28.35kunigami<PROTECTED>
13:30.31kunigamiI've been reading refract.c but
13:31.04kunigamiI couldn't notice any difference, except that it changes the hit callback.
13:34.50kunigamiIn osl_rt, first I detect the ray is a refraction one (which occurs when the next ray to be shot and the normal vector -- which I assume is pointing outwards -- have dot product less than zero). After shooting this new ray, I invert the normal given by the application.
13:35.55brlcadstarseeker: I wouldn't think so, if we pass all release testing -- just have to make sure everything really works with actual hands-on modeling
13:37.06brlcadkunigami: if you're inside an object, you won't get a hit until you encounter another object exterior
13:37.19brlcadi believe you'll want to back the ray out of anything that you're in
13:37.55brlcadat least for onehit
13:38.19starseeker/src/libged/draw.c
13:38.20starseeker{standard input}:13976:non-relocatable subtraction expression, "_dgo_open_tcl" minus "L00000000002$pb"
13:39.38brlcadunclean build?  that was just a function that was renamed from dgo_open to dgo_open_tcl
13:40.28brlcadnot used in draw.c, so a lil mystery there
13:41.41kunigamibrlcad: I'm not sure if I understand what you said. When a ray hits a glass object, I need to shoot a refraction ray. This ray will hit this same glass object from inside. I think I need to pass this "internal" ray to OSL system.
13:42.12starseekerah, OK
13:42.19starseekermutters under his breath about autotools...
13:44.21CIA-62BRL-CAD: 03brlcad * r45297 10/brlcad/trunk/src/libged/draw.c: remove the dgo references in here since this file no longer has anything to do with the dgo interface.
13:45.59brlcadkunigami: so in that case, your incoming ray hits glass, you determine that you need to refract, so you don't need to "rehit" that glass surface (so you don't need to back the ray out) -- you just fire from the interior
13:46.23brlcadyou may need to move the ray forward just an epsilon so that it doesn't accidentically rehit the surface due to floating point fuzz
13:49.33CIA-62BRL-CAD: 03brlcad * r45298 10/brlcad/trunk/src/libged/bev.c: replace the ged_ prefix on non-public API
13:52.35kunigamibrlcad: I didn't considered that! Let me try moving the starting point a little bit in the ray direction.
13:53.35CIA-62BRL-CAD: 03starseeker * r45299 10/brlcad/trunk/src/other/libpng/CMakeLists.txt: This symlink probably shouldn't be pointing to the build directory - point it to the final installed location and see if that works...
13:53.53CIA-62BRL-CAD: 03brlcad * r45300 10/brlcad/trunk/src/libged/bot_dump.c: more ged_ prefix removal on non-public API
14:08.59*** join/#brlcad _psilva (~silvap@static-96-255-52-7.washdc.fios.verizon.net)
14:09.10_psilvamornin
14:09.13brlcadhowdy
14:09.33_psilvai'll try svn tonight
14:11.28CIA-62BRL-CAD: 03starseeker * r45301 10/brlcad/trunk/src/other/libpng/CMakeLists.txt: Spoke too soon - might be a problem with how BRL-CAD is setting CMAKE_LIBRARY_OUTPUT_DIRECTORY
14:12.20CIA-62BRL-CAD: 03starseeker * r45302 10/brlcad/trunk/CMakeLists.txt: See whether we actually needed to specify the BRLCAD_BINARY_DIR - if we don't, this may avoid libpng's problem.
14:20.00CIA-62BRL-CAD: 03starseeker * r45303 10/brlcad/trunk/CMakeLists.txt: Ah, that's right - that's what gives us the ability to duplicate the installed layout. Will need to do something else with libpng.
14:25.45starseekerdg_obj.c:339: warning: implicit declaration on vdraw_cmd_t
14:29.12_psilvai found g_var still in the distro
14:29.22_psilvadoubt it's used by anyone
14:35.33CIA-62BRL-CAD: 03brlcad * r45304 10/brlcad/trunk/src/libged/ (clip.c color.c ged_private.h): more ged_ (and _ged_) prefix removal. great example where reordering eliminates the need for forward decls and the unnecessary inclusion of the color funcs in ged_private (implying internal reuse API).
14:36.04starseekerbrlcad:  are you able to get a build of BRL-CAD right now?  I can't even in non-strict mode...
14:36.18starseekersrc/libged/vdraw.c:740: error: conflicting types for ‘vdraw_cmd’
14:36.30starseekerinclude/dg.h:264: error: previous declaration of ‘vdraw_cmd’ was here
14:39.38CIA-62BRL-CAD: 03brlcad * r45305 10/brlcad/trunk/src/libged/comb_std.c: ged_/GED_ prefix removal on non-public API
14:40.07brlcadstarseeker: I do get a clean build, lemme see if I have uncommitted files
14:41.24CIA-62BRL-CAD: 03brlcad * r45306 10/brlcad/trunk/include/dg.h: sure enough, uncommitted file. vdraw_cmd() was renamed to vdraw_cmd_tcl().
14:42.24brlcad_psilva: it's not been a maintenance burden, so it's been left alone
14:43.02brlcaddead code tends to get left alone unless/until it incurs a maintenance cost
14:44.16starseekerbrlcad: ah, phew - thanks
14:44.26starseeker(kinda makes release testing harder :-P)
14:45.29_psilvajust saying practically it really has no use
14:45.40_psilvai suppose maybe as a reference for another convertor
14:45.44_psilvait has some value
15:21.19brlcad_psilva: you have the ability to make it useful ;)
15:25.40_psilvaheh, just saying
16:48.59*** join/#brlcad yukonbob (~bch@S0106002191d1591c.ok.shawcable.net)
17:25.48CIA-62BRL-CAD: 03kunigami * r45307 10/brlcad/trunk/src/liboptical/osl_rt.cpp: osl-rt can read number of samples from input
17:52.03CIA-62BRL-CAD: 03starseeker * r45308 10/brlcad/trunk/src/other/libpng/CMakeLists.txt: Try an alternative approach to the symlink issue.
17:55.31brlcad_psilva: just saying you're going to do it?
17:58.41brlcadactually, I recall looking at that converter a few months back and decided to keep it around because it might be useful as a plugin to our geometry conversion library
17:58.41CIA-62BRL-CAD: 03starseeker * r45309 10/brlcad/trunk/src/other/libpng/CMakeLists.txt: Fix up indenting.
18:21.20CIA-62BRL-CAD: 03starseeker * r45310 10/brlcad/trunk/src/other/libpng/CMakeLists.txt: Need to use the location property for the active configuration. Probably other places in BRL-CAD where I need to make this change.
18:35.33CIA-62BRL-CAD: 03starseeker * r45311 10/brlcad/trunk/src/other/libpng/CMakeLists.txt: Make sure we flush any old symlinks when re-running cmake.
18:36.50CIA-62BRL-CAD: 03bhinesley * r45312 10/brlcad/trunk/src/libged/ptranslate.c: after pointer arithmetic, argv[0] doesn't point to command name
18:48.27CIA-62BRL-CAD: 03starseeker * r45313 10/brlcad/trunk/src/other/libpng/CMakeLists.txt:
18:48.27CIA-62BRL-CAD: Since Windows doesn't symlink, we need to do something else. Rather than run
18:48.28CIA-62BRL-CAD: code on installation, just add a target to do the copy that depends on the
18:48.28CIA-62BRL-CAD: library target. Can do this for the symlink if need but, but so far it doesn't
18:48.28CIA-62BRL-CAD: look like we need to. Untested on Windows.
18:52.07*** join/#brlcad merzo (~merzo@226-207-132-95.pool.ukrtel.net)
18:55.26CIA-62BRL-CAD: 0399.125.86.110 07http://brlcad.org * r2940 10/wiki/User:Bhinesley: /* Log */ real oed command requires a keypoint
19:00.50CIA-62BRL-CAD: 03starseeker * r45314 10/brlcad/trunk/CMakeLists.txt: LOCATION -> LOCATION_
19:15.05CIA-62BRL-CAD: 03starseeker * r45315 10/brlcad/trunk/CMakeLists.txt:
19:15.05CIA-62BRL-CAD: Most of the time if you're compiling you want Debug build, and when you don't
19:15.05CIA-62BRL-CAD: want that you're specifying Release build - it's extremely rare to really WANT
19:15.05CIA-62BRL-CAD: to not have the build type set. Make the default behavior the common case.
19:27.47``Erikhttp://www.youtube.com/watch?v=X9eriClHWLw  adam savage singing "I will survive" as gollum
19:34.58CIA-62BRL-CAD: 03bhinesley * r45316 10/brlcad/trunk/src/libged/translate.c: Enable support for translating all instances of a primitive. Negative coordinates no longer mistaken for arguments.
19:45.22starseekerbites back cuss words - CMake has it's own notion of permissions, independent of umask
20:44.15*** join/#brlcad bhinesley (~bhinesley@adsl-99-125-86-110.dsl.bkfd14.sbcglobal.net)
20:48.32starseekerbhinesley: I'm seeing the following:
20:48.34starseeker../../../src/libged/translate.c: In function 'ged_translate':
20:48.34starseeker../../../src/libged/translate.c:237: error: expected ')' before 'restrict'
20:48.49bhinesleyhuh, ok
20:50.45*** join/#brlcad yukonbob (~bch@S0106002191d1591c.ok.shawcable.net)
20:53.38CIA-62BRL-CAD: 03bhinesley * r45317 10/brlcad/trunk/src/libged/translate.c: restrict qualifier causing build errors
20:54.06bhinesleyI'm rebuilding from scratch right now, so I haven't tested that, but it should get you by (possibly with a warning)
21:18.57CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2941 10/wiki/ESA_Summer_of_Code_in_Space: initial description of SOCIS announcing intent to apply
21:22.26CIA-62BRL-CAD: 03kunigami * r45318 10/brlcad/trunk/src/liboptical/ (7 files): changed osl-renderer to liboslrend
21:30.15kunigamiI can't get refraction to work. In the following render, the tall box is from glass : http://dl.dropbox.com/u/1399996/GSoC/OSL_RT-2011-06-29.png
21:40.09CIA-62BRL-CAD: 03r_weiss * r45319 10/brlcad/trunk/src/librt/primitives/nmg/nmg_bool.c:
21:40.09CIA-62BRL-CAD: Improved function 'nmg_bool' within file 'nmg_bool.c'. These changes will, under
21:40.09CIA-62BRL-CAD: certain conditions, increase the speed in which boolean operations are performed
21:40.09CIA-62BRL-CAD: on nmg structures. This will increase the speed of operations such as when using
21:40.09CIA-62BRL-CAD: the mged 'facetize' or 'ev' command.
21:49.56brlcadkunigami1: would be a good question to pose to the mailing list, maybe see if Rossberg has some suggestions
21:50.36brlcadyou're definitely getting some interesting effects, but reflectivity is clearly dominant
22:13.16CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2942 10/wiki/ESA_Summer_of_Code_in_Space: simplify
22:18.00CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2943 10/wiki/ESA_Summer_of_Code_in_Space: link proj ideas, needs customization to socis
22:19.24CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/move: [[Google Summer of Code/Checklist]] moved to [[Summer of Code/Checklist]]: want to reuse the checklist since ESA is running their own SoC
22:24.26CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2946 10/wiki/Summer_of_Code/Checklist: generalize and specify
22:25.37CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/move: [[Google Summer of Code/Application Guidelines]] moved to [[Summer of Code/Application Guidelines]]: Now supports two SoC programs
22:26.30CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2949 10/wiki/Summer_of_Code/Application_Guidelines: socify
22:27.01CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2950 10/wiki/Summer_of_Code/Checklist: simplify category
22:27.46CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/move: [[Google Summer of Code/Acceptance]] moved to [[Summer of Code/Acceptance]]: now supporting two soc programs
22:32.05CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2953 10/wiki/Summer_of_Code/Acceptance: generalize
22:32.56CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/move: [[Google Summer of Code/Expectations]] moved to [[Summer of Code/Expectations]]: now supporting two SoC programs
22:37.12CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2956 10/wiki/Summer_of_Code/Expectations: generalize
22:37.51CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2957 10/wiki/Summer_of_Code/Checklist: be abnoxious
22:38.53CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2958 10/wiki/Summer_of_Code/Checklist: or not
22:40.26CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2959 10/wiki/Google_Summer_of_Code:
22:41.38CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2960 10/wiki/Google_Summer_of_Code/2011:
22:59.55CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2961 10/wiki/ESA_Summer_of_Code_in_Space: generalize
23:01.59CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/delete: deleted "[[Category:Google Summer of Code]]": content was: '[[category:Projects]]' (and the only contributor was '[[Special:Contributions/Ssd|Ssd]]')
23:02.53CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2962 10/wiki/Google_Summer_of_Code/2008:
23:03.26CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r0 10/wiki/Special:Log/move: [[Google Summer of Code/Proposal Evaluation]] moved to [[Summer of Code/Proposal Evaluation]]: generalize
23:05.04CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2965 10/wiki/Summer_of_Code/Proposal_Evaluation: generalify
23:05.50CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2966 10/wiki/User:EBautu: cat
23:06.33CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2967 10/wiki/More_Changelog: cat
23:07.00CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2968 10/wiki/Google_Summer_of_Code/Flyers:
23:07.27CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2969 10/wiki/Google_Summer_of_Code/2009/Project_Ideas: cat
23:07.50CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2970 10/wiki/Google_Summer_of_Code/2008/Project_Ideas: cat
23:08.01CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2971 10/wiki/Google_Summer_of_Code/2009: cat
23:08.27CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2972 10/wiki/Google_Summer_of_Code/2009/Project_Ideas: stray .
23:11.08CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2973 10/wiki/Google_Summer_of_Code/Project_Ideas: generalize
23:14.19CIA-62BRL-CAD: 03Sean 07http://brlcad.org * r2974 10/wiki/ESA_Summer_of_Code_in_Space/Project_Ideas: initial stab at ideas for SOCIS, starting with the GSoC list
23:59.32``Erik*grouse* I might have to play a game with highlighters and a trip to someone with a free tube tester O.o

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