IRC log for #brlcad on 20080619

00:20.23 *** join/#brlcad Twingy (n=justin@74.92.144.217)
00:30.44 *** join/#brlcad PrezKennedy (i=Matthew@whitecalf.net)
00:31.43 *** join/#brlcad andre|away (n=chatzill@189.71.51.90)
02:21.19 yukonbob hello, cadhead
02:21.21 yukonbob *s
02:21.32 pacman87 hi yukonbob
02:21.41 yukonbob hey pacman87, how're tricks?
02:21.51 pacman87 tricks?
02:22.19 yukonbob *How're things going?
02:23.04 pacman87 pretty good, hyp is just about finished, and i'm starting to research for rev/sweep
02:23.16 pacman87 you?
02:24.07 yukonbob oh... I'm tuckered out. Long hours... no time to geek out w/ BRL-CAD, unfortunately... and I _need_ to so I can get the latest revs building/running nicely...
02:24.20 yukonbob but I'm alive, so that's good.
02:24.31 yukonbob (some might disagree)
02:30.42 dtidrow heh
02:33.58 *** join/#brlcad Tony__ (n=chatzill@ppp035.cri.centurytel.net)
05:49.40 *** join/#brlcad clock_ (n=clock@217-162-109-217.dclient.hispeed.ch)
06:58.46 homovulgaris wakes up after 9 hours of sleep
06:58.56 homovulgaris that was a long nap :P
06:59.20 *** join/#brlcad clock_ (n=clock@zux221-122-143.adsl.green.ch)
07:03.26 brlcad heh
07:15.04 homovulgaris hi sean, didnt see rossberg
07:15.21 brlcad ~seen rossberg
07:15.24 ibot rossberg <n=rossberg@bz.bzflag.bz> was last seen on IRC in channel #brlcad, 448d 16h 32m 32s ago, saying: 'brlcad: sorry, "fmax" is missing on ftp.brlcad.org'.
07:15.29 brlcad ~seen drossberg
07:15.30 ibot brlcad: i haven't seen 'drossberg'
07:15.33 brlcad gr
07:15.40 homovulgaris :D
07:15.48 brlcad ah
07:15.52 brlcad ~seen d_rossberg
07:15.53 ibot d_rossberg <n=rossberg@bz.bzflag.bz> was last seen on IRC in channel #brlcad, 1d 19h 51m 39s ago, saying: ' (but not all ;-)'.
07:16.28 brlcad homovulgaris: it was more when you get the chance to ask him
07:17.14 homovulgaris :) btw, i was planning on using this for the graph representation : http://www.boost.org/doc/libs/1_35_0/libs/graph/doc/adjacency_list.html
07:17.31 homovulgaris but i have to check about defining graph traversal system
07:18.55 homovulgaris i hope to be able to make a simple constraint solution search system in the next few days: brute force with simple backtracking
07:19.26 brlcad hm, you'd think we'd have a graph structure already
07:19.38 brlcad given the CSG hierarchy is a directed acylcic graph
07:20.42 homovulgaris hmm.. should check that out then .. would the structure definition be in bu ?
07:20.42 brlcad you just might be "in luck" though, I'm not seeing or remembering a specific structure that would help other than our red-black trees (and that's a stretch)
07:21.09 brlcad bu, bn, or rt
07:21.15 brlcad maybe to a quick search
07:21.22 brlcad but I'm not remembering
07:21.28 brlcad good question for daytona if you see him
07:21.45 brlcad though I think the answer is still no
07:21.53 homovulgaris k. i'll mail him , update wiki
07:22.15 brlcad good idea, copy the devel list
07:22.24 brlcad (if you mail him)
07:22.43 homovulgaris and the idea of rt_params was to just get the set of parameters.. constraint return would be via pc_constraint_import
07:23.21 brlcad I meant implicit constraints that all primitives have
07:23.36 brlcad like how a sphere has implicit constraints that a==b==c
07:23.39 homovulgaris to set the parameters i was thinking of adding another function.. but that seems unnecessary.. i should integrate it to params itself ?
07:23.43 brlcad where a,b,c are the params
07:23.49 homovulgaris yeah implicit constraints ofcourse
07:24.06 brlcad hm?
07:24.07 homovulgaris i mean implicit constraints would be returned by rt_*_params
07:24.13 brlcad ah, okay
07:24.21 brlcad so some way to delineate between the two
07:25.33 homovulgaris two ? u mean set / return ?
07:25.49 brlcad yeah, the set being returned
07:26.24 homovulgaris hey sean, what is the convention .. right now i have all the method definitions for the classes inside .h files .. should i shift them to .cpps ?
07:26.30 homovulgaris with one file for each class ?
07:26.35 brlcad i mean you're either making two functions, one for parameters and one for the primitive constraints, or one func that returns both with a means to distinguish which is a parameter and which is a constraint
07:27.43 homovulgaris the idea of params() right now is to return both parameters and constraints while distinguishing between both..
07:27.44 brlcad homovulgaris: that's not so much a matter of convention as it is intent -- if you don't intend them to be inlined, they should probably go into a cpp
07:28.04 brlcad and the .h then just describes and documents the class/methds
07:28.18 brlcad okay, that's what I though
07:28.19 homovulgaris k
07:28.31 brlcad params() might not be the best name, but I don't have another suggestion :)
07:28.55 homovulgaris that can always be changed :) i suck at naming.. even pc sounds crappy ;)
07:29.14 brlcad you should try to keep it to one class per file, unless they're a bunch of really small related classes
07:29.43 brlcad pc works surprisingly well
07:30.02 homovulgaris hmm. right now that .h is around 350 lines without comments.. so i was thinking split it into 2
07:30.05 brlcad and the namespace conflicts on libpc are similarly conveniently very minimal
07:30.22 brlcad "that .h"?
07:30.29 homovulgaris pc_solver.h
07:31.18 homovulgaris will be back after lunch :)
07:31.22 brlcad I'd imagine it would necessarily break up
07:31.35 brlcad e.g. the generic interval class kinda belongs by itself
07:31.48 homovulgaris yeah.. that was my plan
07:31.58 brlcad same for domain, etc
07:32.05 brlcad you've got a lot of classes in there
07:32.11 homovulgaris probably Interval in one.. and Domain and variable together or 3 separate..
07:32.23 homovulgaris some of them have to be removed / upgraded :D
07:32.29 homovulgaris i mean the empty ones :)
07:32.42 brlcad try to keep it to one unless they really are so tightly coupled to each other that they don't make sense without the other
07:32.59 brlcad or if one is effectively just a private class that another class uses
07:33.14 brlcad (akin to a static struct)
07:33.31 homovulgaris hmm not exactly i think in the near future we will have inheritances from Interval for example
07:34.01 brlcad the inherited classes become domain-specific
07:34.03 homovulgaris for discrete intervals or something like that or from Variable for different types
07:34.08 brlcad that alone is a good reason to separate
07:34.16 homovulgaris yeah
07:34.19 brlcad if these were broken into sub-libs, Interval is a generic concept
07:34.51 brlcad it'd be in some base common library because it's application/domain agnostic
07:34.59 homovulgaris I was thinking of adding some more functionality into interval later.. it is a pretty useful concept..
07:35.01 brlcad you only have one lib, so it just gets its own file
07:35.11 homovulgaris ok
07:35.39 homovulgaris after the minimal implementation of constraint solution search i will work on the stack system for handling constraint representation
07:35.59 brlcad if you want to extend something like a BoundingBoxInterval class or something, he'd get his own implementation/header files
07:36.28 homovulgaris and when i am stuck with C++ i can add stuff to each of the params/ expand the pc_import etc.
07:36.42 homovulgaris ok.
07:37.24 homovulgaris and sean, regarding the regression testing on the roots change .. i was confused.. am i supposed to do the testing :) ?
07:37.25 brlcad yeah, hooking into a given primitive to use params() and your solver in import or prep would be a really good test of the system
07:38.05 brlcad homovulgaris: no, don't worry about that -- focus on libpc
07:38.23 brlcad i was going to do the testing
07:38.27 homovulgaris i just want to make sure that the whole architecture works for the simple case and then add on features
07:38.36 homovulgaris that was about lipbc
07:38.43 brlcad you're wouldn't necessarily know exactly what to look for anyways
07:39.01 brlcad since it's the portions that relate to ray-trace behaviors that are of concern
07:39.21 homovulgaris yeah.. i was thinking about it.. i mean i have 0 knowledge about regression testing.. :) will definitely need to check it out over the year though
07:42.36 homovulgaris also presently the Interval class doesnt have sorting depending on optimal nature.. It just sorts according to < operator definition. So if i need to change it in a derived class i can redefine the < operator or write a sort function.
07:43.53 homovulgaris which i we will need when we do the solution search selectively over more preferred intervals or to represent preferred solutions
07:44.09 brlcad nods
07:44.13 brlcad sounds reasonable
07:44.48 homovulgaris ok... i'd be back after lunch
07:44.57 brlcad happy chowda
07:49.56 *** join/#brlcad shazbot (i=6152a301@gateway/web/ajax/mibbit.com/x-f3e19621aa163273)
07:50.44 shazbot Hi, just heard about the google summer of code.. good luck. Can anyone tell me what the lgpl-ed parts of BRL-cad are?
07:54.34 brlcad shazbot: almost the entire source code is lgpl
07:55.11 brlcad COPYING has the details
07:55.57 brlcad source code is lgpl unless it's not ours to license; build system, documentation, and data files are mostly bsd licensed
08:09.01 *** join/#brlcad vector_crazy (i=6152a301@gateway/web/ajax/mibbit.com/x-ec5797b3af2023dc)
08:36.59 homovulgaris is back :)
08:37.17 homovulgaris loves chicken
08:39.46 *** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6)
08:54.15 *** join/#brlcad andrecastelo (n=chatzill@189.71.51.90)
09:22.53 *** join/#brlcad mafm (n=mafm@elnet-111.lip.pt)
09:24.05 mafm allo
09:38.08 CIA-22 BRL-CAD: 03andrecastelo * r31465 10/brlcad/trunk/src/rt/viewmlt.c:
09:38.08 CIA-22 BRL-CAD: Added default lights and scanline initializations in view_2init(). Added
09:38.08 CIA-22 BRL-CAD: scanline freeing in view_end(). Added view_cleanup() function. These functions
09:38.08 CIA-22 BRL-CAD: are pretty similar to rt's, the big differences will come in rayhit() and in
09:38.08 CIA-22 BRL-CAD: view_pixel().
12:18.22 *** join/#brlcad elite01 (n=elite01@dslb-088-071-045-101.pools.arcor-ip.net)
12:39.15 *** join/#brlcad d_rossberg (n=rossberg@bz.bzflag.bz)
12:39.40 *** join/#brlcad vedge (n=vedge@205-237-251-204.ilesdelamadeleine.ca)
13:13.19 *** join/#brlcad thing0 (n=ric@58.171.188.128)
13:17.15 CIA-22 BRL-CAD: 03bob1961 * r31466 10/brlcad/trunk/src/libged/ (CMakeLists.txt Makefile.am): Add more commands.
13:24.35 CIA-22 BRL-CAD: 03bob1961 * r31467 10/brlcad/trunk/src/libged/ (11 files): Initial check-in.
13:27.53 CIA-22 BRL-CAD: 03bob1961 * r31468 10/brlcad/trunk/src/libged/ (erase.c erase_all.c illum.c): Initial check-in.
13:34.45 CIA-22 BRL-CAD: 03bob1961 * r31469 10/brlcad/trunk/src/libged/globals.c: Added FreeSolid.
14:03.08 CIA-22 BRL-CAD: 03bob1961 * r31470 10/brlcad/trunk/src/libged/bigE.c: Converted to use struct ged.
14:04.27 CIA-22 BRL-CAD: 03bob1961 * r31471 10/brlcad/trunk/src/libged/ged_private.h: Added more function declarations and a few struct definitions.
14:10.31 CIA-22 BRL-CAD: 03bob1961 * r31472 10/brlcad/trunk/src/libged/draw.c: Added ged_draw_guts() and ged_ev().
14:11.46 CIA-22 BRL-CAD: 03bob1961 * r31473 10/brlcad/trunk/src/libged/dg_obj.c: Temporarily disable the E command.
14:14.45 CIA-22 BRL-CAD: 03bob1961 * r31474 10/brlcad/trunk/include/ged.h: Added declarations for several functions.
14:15.39 CIA-22 BRL-CAD: 03bob1961 * r31475 10/brlcad/trunk/misc/win32-msvc8/libged/libged.vcproj: Added several source files.
14:18.00 CIA-22 BRL-CAD: 03brlcad * r31476 10/brlcad/trunk/HACKING: update how to make the ChangeLog
14:19.09 CIA-22 BRL-CAD: 03bob1961 * r31477 10/brlcad/trunk/src/libged/ged_private.h: Added include for time.h
14:21.43 CIA-22 BRL-CAD: 03brlcad * r31478 10/brlcad/trunk/src/libged/ged_private.h: common before system header consistency
14:24.04 pacman87 morning, all
14:24.05 *** join/#brlcad Elperion (n=Bary@p5B14E854.dip.t-dialin.net)
14:40.34 *** join/#brlcad madant (n=homovulg@202.63.233.61)
14:43.23 starseeker go bob go!
15:13.16 *** join/#brlcad vedge (n=vedge@205-237-251-204.ilesdelamadeleine.ca)
15:38.25 *** join/#brlcad homovulgaris (n=homovulg@202.63.233.61)
15:43.29 CIA-22 BRL-CAD: 03starseeker * r31479 10/brlcad/trunk/doc/docbook/ (5 files in 3 dirs): (log message trimmed)
15:43.30 CIA-22 BRL-CAD: Proof of concept - the new mged_commands.xml contains a little of the extensive
15:43.30 CIA-22 BRL-CAD: command appendix in Volume II as a valid stand-alone article, and the
15:43.30 CIA-22 BRL-CAD: VolumeII.xml here references inside that file to grab the content for inclusion
15:43.31 CIA-22 BRL-CAD: in its own appendix using XInclude and xpointer. If this technique scales, it
15:43.33 CIA-22 BRL-CAD: will be possible to break down the volumes into individual valid files that can
15:43.35 CIA-22 BRL-CAD: be assembled by VolumeII.xml to reproduce (more or less) the original VolumeII
15:53.09 CIA-22 BRL-CAD: 03bob1961 * r31480 10/brlcad/trunk/ (include/bu.h src/libbu/argv.c): Added bu_argv_from_path. bu_copy_argv and bu_copyinsert_argv changed to bu_dup_argv and bu_dupinsert_argv, respectively.
15:54.04 CIA-22 BRL-CAD: 03bob1961 * r31481 10/brlcad/trunk/src/mged/rtif.c: bu_copyinsert_argv changed to bu_dupinsert_argv
15:54.59 CIA-22 BRL-CAD: 03bob1961 * r31482 10/brlcad/trunk/src/tclscripts/helplib.tcl: Changed help for dgo_E.
16:42.58 CIA-22 BRL-CAD: 03starseeker * r31483 10/brlcad/trunk/doc/docbook/system/mged/mged_commands.xml: 2 command explanations and examples added, using glossary docbook markup
16:59.14 CIA-22 BRL-CAD: 03starseeker * r31484 10/brlcad/trunk/doc/docbook/system/mged/mged_commands.xml: More prep work for glossary, another example
17:06.40 *** join/#brlcad louipc (n=louipc@76-10-146-181.dsl.teksavvy.com)
17:10.32 poolio brlcad: Does it make sense that arb8s are rt_arb and not arb8?
17:34.51 CIA-22 BRL-CAD: 03mafm * r31485 10/rt^3/trunk/src/g3d/History.h: Adding header with copyright and so on; and added missing #endif preventing the program from compiling successfully.
17:42.14 brlcad poolio: probably not
17:42.55 brlcad arb8 used to be g_arb.c -- but for consistency they were all made to match their functab name
17:43.20 brlcad for which arb8 actually handles arb8, arb7, arb6, arb5, and arb4
17:49.33 poolio ah oh well. I've decided to work on arb8 instead of arbn :)
17:49.57 poolio You've already done most the arb8 work...just need to adjust for equivalent vertices
17:55.15 brlcad yeah, the arb8 nmg routine is already in brep form
17:55.28 brlcad notes a distinct lack of commits if progress is being made :)
17:56.50 poolio brlcad: so calling rt_arb_tess() is probably a good idea here?
17:57.13 brlcad up to you :)
17:57.26 poolio Also, there are a few places where primites are mentioned as say, "g_arb.c" ... should those be changed to arb8.c?
17:57.26 brlcad if you just end up with nearly the same logic, probably
17:57.35 brlcad sure
17:58.00 poolio Ah k, I will commit when it compiles :)
17:58.31 brlcad even better would be to have one "structure" routine for brep/nurb/tess if they're basically the same
17:59.37 poolio as in something that went from NMG -> brep?
18:00.47 poolio Ah, the one big question I still haven't resolved is how to properly iterate over elements of the NMG. It looks like repeatedly calling the GET routines does it?
18:01.29 brlcad something like that -- basically just so there's not two or three copies of logic that do effectively the same thing
18:01.56 brlcad mm, I forget how to deal with nmgs
18:02.04 brlcad it involves loops
18:02.08 brlcad lots of loops
18:02.44 poolio ah hmm, I couldn't find anywhere in the code where they repeatedly called GET, but I didn't see any other way to really access them
18:03.31 brlcad almost all the converters iterate over nmgs, could check their code out in src/conv
18:03.37 poolio ah, thanks
18:04.06 brlcad the E and ev commands too
18:04.45 poolio Ah hey, I see...it's all fun linked list goodness :)
18:06.00 brlcad yep
18:11.31 mafm brlcad: so should I commit all the code of my libraries to src/other of my module, right?
18:12.00 brlcad mafm: you mean the three dependencies?
18:12.09 mafm 4
18:12.19 brlcad oh? who's number 4?
18:12.19 mafm ogre, ois, mocha, rbgui
18:12.28 brlcad ah, ois isn't part of ogre?
18:12.49 brlcad but yeah, to answer your question
18:12.49 mafm they decided to remove input support from OGRE about 2 years ago
18:13.10 brlcad the first commit should be a clean upstream
18:13.15 brlcad i.e. no patches/mods
18:13.19 brlcad then commit the mods
18:13.35 brlcad so it'll be easier to extract/reapply later if needed as well as to see what changes there are
18:14.08 mafm I got a tarball for OIS, but for OGRE has to be some specific revision of the trunk that happened to work a while ago -- today's isn't even compiling :D
18:17.20 brlcad awesome
18:17.32 brlcad any upstream revision, even from svn/cvs/git/whatever is fine
18:17.55 brlcad just annotate their revision in the commit message (e.g. from upstream ogre svn repository, r12345)
18:28.15 CIA-22 BRL-CAD: 03mafm * r31486 10/rt^3/trunk/src/other/ (. Makefile): Commiting other/ subdir and the Makefile
18:29.09 CIA-22 BRL-CAD: 03brlcad * r31487 10/brlcad/trunk/TODO:
18:29.09 CIA-22 BRL-CAD: next release is still a fair bit away given the libged instability, but go ahead
18:29.09 CIA-22 BRL-CAD: and annotate some of the most pressing issues for the next release as testing
18:29.09 CIA-22 BRL-CAD: mged, verifying fast4-g for ajem, and fixing the bug tom browder just reported
18:29.09 CIA-22 BRL-CAD: about the windows installer
18:32.40 CIA-22 BRL-CAD: 03brlcad * r31488 10/brlcad/trunk/NEWS: note that dawn added a -c option to the loop command so you can now iterate over characters as well as numbers. makes loop even more similar to the old bsd 'jot' command, but with a better name.
18:36.47 brlcad homovulgaris: fyi, your r31454 commit blew away my changes
18:37.12 mafm sigh, propsets...
18:37.26 brlcad mafm: sup?
18:42.59 mafm it's missing like 30 propsets only in ois/ :)
18:46.02 mafm hmm, scripts are text/plain, right? xml have to be text/xml or can be text/plain too?
18:46.52 starseeker feels mafm's pain
18:47.02 starseeker xml should probably be text/xml
18:48.09 mafm in my case they don't even name the file .xml
18:48.26 mafm it's some random msvc/mac stuff :P
18:55.12 brlcad heh
18:55.45 brlcad text/plain or test/xml is fine for their stuff
18:55.51 brlcad they don't matter as much as our stuff
18:56.41 mafm hmmm, and OGRE now doesn't want to build with the old version either, great >:|
18:57.30 mafm svn propset svn:eol-style native ois/src/linux/LinuxKeyboard.cpp
18:57.30 mafm svn: File 'ois/src/linux/LinuxKeyboard.cpp' has inconsistent newlines
18:57.30 mafm svn: Inconsistent line ending style
18:57.56 alex_joni run dos2unix on it
18:58.25 mafm it seems so, now works :)
19:03.10 CIA-22 BRL-CAD: 03mafm * r31489 10/rt^3/trunk/src/other/ois/ (155 files in 29 dirs): Commiting OIS (Object-Oriented Input System) from tarball v1.2.0, no patches needed
19:03.27 mafm yay, at last
19:19.48 mafm now for OGRE it'll need a few iterations and many minutes of transmissions, I go to watch the football match to the public square
19:19.52 mafm :D
19:20.25 mafm maybe I won't come back if it's too late, so see you tomorrow then
19:23.53 *** join/#brlcad clock_ (n=clock@77-56-94-137.dclient.hispeed.ch)
19:25.23 poolio brlcad: Is there a way to get a list of all vertices in an NMG model? The only way I can think of doing it is going through all the vertexuses
19:25.58 *** join/#brlcad Elperion (n=Bary@p5B14E854.dip.t-dialin.net)
19:26.30 CIA-22 BRL-CAD: 03bob1961 * r31490 10/brlcad/trunk/src/libged/ (ged.c how.c set_transparency.c tree.c): Initial check-in.
19:28.35 CIA-22 BRL-CAD: 03bob1961 * r31491 10/brlcad/trunk/ (include/ged.h src/libged/qray.c): Changed signature of ged_init_qray(). Added declaration of ged_free_qray.
19:29.53 CIA-22 BRL-CAD: 03bob1961 * r31492 10/brlcad/trunk/src/libged/nirt.c: Mods related to signature change of ged_qray_data_to_vlist.
19:31.52 CIA-22 BRL-CAD: 03brlcad * r31493 10/brlcad/trunk/doc/ (mged.sh mged.tr): renamed to .sh since it's a shell script archive
19:31.58 CIA-22 BRL-CAD: 03bob1961 * r31494 10/brlcad/trunk/src/libged/nirt.c: Oops! Forgot to change the usage :-).
19:47.42 CIA-22 BRL-CAD: 03bob1961 * r31495 10/brlcad/trunk/src/libged/ged.c: Fix miscellaneous descrepancies. Note - this has still not been tested.
19:51.25 *** join/#brlcad madant (n=homovulg@202.63.233.61)
19:53.36 CIA-22 BRL-CAD: 03bob1961 * r31496 10/brlcad/trunk/src/libged/ (nirt.c qray.c): After actually compiling and noticing the need for ged_wdbp->dbip the ged_qray_data_to_vlist function has been changed back to using a struct ged.
19:55.55 CIA-22 BRL-CAD: 03brlcad * r31497 10/brlcad/trunk/doc/Makefile.am: mged.tr was renamed to mged.sh
20:18.38 brlcad madant: fyi, your r31454 commit blew away my changes .. (if you missed the earlier note)
20:22.48 PrezKennedy why arent you partying??
20:23.49 *** join/#brlcad homovulgaris (n=homovulg@202.63.233.61)
20:23.58 CIA-22 BRL-CAD: 03bob1961 * r31498 10/brlcad/trunk/ (4 files in 3 dirs): added wdb_init()
20:27.31 homovulgaris brlcad: sorry.. u mean the ones in solver_test.c ?
20:28.22 *** part/#brlcad louipc (n=louipc@76-10-146-181.dsl.teksavvy.com)
20:29.06 homovulgaris the code indentation right ? hmm.. wait or was it pc_test.. yikes sorry there was a conflict i guess i resolved it wrongly
20:29.55 homovulgaris but i didnt commit that in 31454 .. argh.. am confused ..
21:30.58 *** join/#brlcad elite01 (n=elite01@dslb-088-071-045-101.pools.arcor-ip.net)
21:49.04 *** join/#brlcad dtidrow (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net)
22:57.04 *** join/#brlcad dtidrow (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net)
23:12.26 *** join/#brlcad madant (n=homovulg@202.63.233.61)

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