IRC log for #brlcad on 20080624

00:24.18 *** join/#brlcad houdin8888 (n=houdin88@ip24-250-205-120.ga.at.cox.net)
00:25.36 CIA-22 BRL-CAD: 03starseeker * r31574 10/brlcad/trunk/doc/docbook/ (11 files in 3 dirs): Add Lesson 16
00:25.46 starseeker Wooooo-hooooooo!
00:26.54 houdin8888 Hi all. I'm new to this BRL-CAD. Just wanted to say hello.
00:27.08 starseeker peels eyeballs off monitor
00:27.13 pacman87 houdin8888: hi
00:27.19 starseeker howdy and welcome
00:27.30 houdin8888 thank you. thank you
00:28.24 houdin8888 just trying to get me teeth into some CAD experience. thought i'd go the open-source route. glad there's so much out there.
00:29.46 houdin8888 would anyone suggest an easy tutorial for someone (myself) wishing to start using BRL-CAD on MS Vista?
00:32.32 houdin8888 i wanted to try the "benchmark" images using script run.sh, but couldn't find the "bench" directory.
00:34.31 houdin8888 hmm. did i say something wrong?
00:35.58 houdin8888 ?lib
00:39.41 starseeker heh - no
00:39.44 starseeker we're in and out
00:40.37 houdin8888 ic, its ok. i know i have a very high learning curve to climb.
00:40.55 starseeker For tutorials the standard starting point is here: http://brlcad.org/w/images/c/cf/Introduction_to_MGED.pdf
00:41.30 houdin8888 ahh.. yes,. I have that one. thanks.
00:42.19 houdin8888 i read the first pdf in the series... haven't yet digested this 2nd pdf
00:42.53 starseeker The second one is the "real" introductory documentation
00:43.07 starseeker in the sense that it is the volume that gets you working with CAD geometry
00:43.09 houdin8888 i'm beginning to see that.
00:43.47 yukonbob hello, cadheads
00:43.55 starseeker Hey yukonbob
00:43.59 houdin8888 Hello!
00:44.02 yukonbob hey starseeker
00:44.30 starseeker is now MUCH closer to having a reasonable Volume II docbook conversion
00:44.34 yukonbob houdin8888: expect a _steep_, steep learning curve, and once you stop and look around, find it's all been worth it ;)
00:44.52 houdin8888 i am very hopeful
00:45.18 yukonbob <yoda>There is no hope; only do, or do not</yoda>
00:45.38 pacman87 there is no try
00:45.38 houdin8888 hehe
00:45.40 starseeker houdin8888: Do you have any experience in other CAD systems?
00:45.41 yukonbob ^--- bastardization, but wth...
00:45.52 houdin8888 googlesketch
00:46.07 yukonbob houdin8888: CSG (ie: POVRay)?
00:46.39 houdin8888 i must admit near total ignorance.. but, been a long time wannabe
00:46.52 starseeker OK. I could go Yoda and say "you must unlearn what you have learned" but it would be more accurate to say that prior experience is not terribly applicable to BRL-CAD
00:47.17 starseeker It's a very DIFFERENT way of modeling for most folks
00:47.35 houdin8888 i like the idea that BRL-CAD was used to predict ballistic effectiveness
00:47.43 starseeker s/was/is
00:47.46 starseeker still in active use
00:48.01 starseeker although BRL-CAD itself doesn't do the military predictions
00:48.01 yukonbob houdin8888: you're getting into the advanced stuff; long path to balistic modelling ;)
00:48.12 starseeker it is used by other software that does
00:48.16 houdin8888 true... but, i like pyrotechnics
00:48.29 starseeker heh - sorry, no explosions
00:48.37 yukonbob predicts somebody is going to be disappointed...
00:49.00 starseeker If you want to animate stuff being blown up, Blender is probably a better bet ;-)
00:49.15 houdin8888 no disappointment.. I'm committed to learning this computer graphics modeling program
00:49.21 yukonbob prefers the easy route and goes to a movie
00:49.24 starseeker can't understand why their second movie project didn't get more notice...
00:49.37 starseeker Big Buck Bunny or something like that
00:49.46 yukonbob hrmm... first I've heard of it...
00:49.55 starseeker one sec...
00:49.57 yukonbob hits alta vista
00:50.15 starseeker www.bigbuckbunny.org
00:51.13 starseeker Not only did they do it with Blender, but all the content files are available under creative commons licenses
00:51.21 yukonbob neat
00:51.40 starseeker figures someone will be doing a bunny vs. penguin faceoff at some point...
00:51.47 yukonbob we're undergoing a mini rennaisance with media and licensing...
00:51.50 starseeker It's near Pixar level animation
00:53.25 starseeker is still annoyed at slashdot for not giving them a front page article
00:53.46 houdin8888 nice find (bigbuckbunny)
00:54.30 starseeker It's the successor project to Elephant's Dream, which was the Blender project's first such move. Well done, very artsy, very weird viewing experience
00:54.55 starseeker Big Buck Bunny was intended to appeal to a wider audience
00:55.20 starseeker Ah, well
00:56.01 starseeker They may have tried to avoid the slashdot effect, given how large the movie files are
00:56.41 starseeker either everyone's watching it or I chased everyone away ;-)
00:56.59 houdin8888 i'm watchin the movie trailer
00:57.04 starseeker heh
00:58.08 starseeker must get out of here, back on later
00:58.20 houdin8888 nice to meet ya
01:12.32 *** join/#brlcad Byron1 (n=byron@pool-96-229-127-10.lsanca.fios.verizon.net)
01:12.56 Byron1 how can I create an empty box with no top
01:13.50 brlcad Byron1: depends what sort of box, thickness of walls, etc
01:14.35 Byron1 I would like to create a box similiar to a battery
01:14.44 brlcad one of the easiest ways is to just create an arb8 (a box), then subtract a slightly smaller arb8 from it that is shifed up
01:15.26 brlcad the 'inside' command will create an inner arb8 easily, then you can just extrude the top face to make it "open"
01:15.37 Byron1 Could the box then not have a top?
01:15.58 Byron1 How can you extrude the top
01:17.09 brlcad ~starseeker++
01:17.13 brlcad nice doc progress!
01:17.54 brlcad Byron1: yes, the same arb8 that makes the exterior can be made to remove the top
01:18.24 brlcad when you're editing an arb8, one of the (many) editing options is moving faces
01:19.18 Byron1 If you move a face can it be removed from the box?
01:19.20 brlcad pacman87: yeah, 'revolve' and 'sweep' probably best -- the old short names were from ancient days where every keystroke was precious and cryptic names were the standard convention
01:20.11 brlcad Byron1: "yes and no" .. you cannot have an arb8 with a side missing, that's topologically not solid geometry
01:20.24 brlcad but you can certainly make a box with a hole in the top
01:20.35 brlcad two primitives, bam
01:22.00 brlcad the geometry must always preserve a solidity constraint, hence solid primitives -- even with brep, it wouldn't be allowed
01:22.36 brlcad Byron1: have you gone through the basic mged tutorials? .. that might help explain things a bit
01:23.18 Byron1 Yes I have gone through the tutorials. And they have helped out a lot.
01:24.29 Byron1 I thought that I could make a glass box with a steal plate inside and be able to see the plate
01:25.17 brlcad pacman87: there is and is not a 3D spline structure you can use.. openNURBS provides it (as do several of our older tools internally like 'track', the anim tools, and joint solving)
01:25.19 Byron1 Also I figured I could do a thin arb8 to make the bottom non transparent
01:25.37 brlcad sounds reasonable
01:26.50 Byron1 if the top was transparent I figured that that would give the impression to the top not there
01:33.17 brlcad it would if it was fully transparent
01:33.52 brlcad the thing is those are region properties, you'd have to create a separate object for each side of the box and make just the top fully transparent .. which would be silly
01:35.14 brlcad makes a quick example
01:42.30 brlcad Byron1: http://brlcad.org/tmp/box.png
01:42.57 brlcad that's four primitives, two that make the glass box (that has no top), one for the bottom checkered plate, and the sphere
01:44.02 brlcad Byron1: and the geometry for that is here: http://brlcad.org/tmp/box.g
01:50.06 brlcad starseeker: you'll see a lot of "very weird viewing experiences" at siggraph, even more so than elephant's dream
01:50.48 brlcad lots of really interesting (and huge) artistic influence.. they tend to be the fun side of siggraph :)
02:08.28 CIA-22 BRL-CAD: 03brlcad * r31575 10/brlcad/trunk/include/vmath.h: add a 2d point type to finish out the set, plus some minor ws consistency updates
02:15.12 CIA-22 BRL-CAD: 03brlcad * r31576 10/brlcad/trunk/include/vmath.h: busted, define the right names so it'll compile
02:16.40 pacman87 brlcad: there was a 2d point, the 2d vector was missing
02:18.43 brlcad just a bad comment
02:18.59 pacman87 just checking ;)
02:19.03 brlcad nods
02:20.56 brlcad also, wouldn't worry about aligning the sketch (does extrude try to?)
02:21.57 brlcad pacman87: also, just background info, it was missing simply because nobody had needed them yet (at least from a vmath perspective)
02:23.19 brlcad try to only add things exactly when they're needed, otherwise maintenance grows faster than is useful if you develop speculatively
02:23.37 brlcad rather, the cost of that maintenance is high
02:24.01 brlcad that goes for most of the code in brl-cad, though some parts less than others (i.e. they need refactoring)
02:24.38 pacman87 aligning the sketch with what?
02:25.24 brlcad you'd asked assuming the y-axis
02:25.37 brlcad or using point/vector
02:26.10 pacman87 right
02:26.35 pacman87 so you're saying just to use the y axis?
02:29.08 brlcad I think that's a safe assumption
02:29.27 brlcad just because, as example, I don't recall extrude accounting for sketch rotation in any way
02:29.32 brlcad not much different
02:30.17 brlcad the only thing I can think of is that it might require some changes to the sketch editor to get something that perfectly clamps to that y-axis
02:30.42 pacman87 for extrude, you give the height vector, and one of the axes (i forget which), and it aligns the sketch with that
02:30.46 brlcad I believe it dynamically scales the domain when the bounds are reached
02:31.33 pacman87 if it helps, i could just ignore everything in the -x side of the sketch
02:31.57 pacman87 so the lines just have to cross the y-axis instead of stopping
02:33.45 brlcad i'm thinking you might just have to auto-adjust the x-domain
02:34.10 brlcad i.e. shift it to the right so it meets the first point/curve
02:34.22 brlcad it == y-axis
02:34.49 pacman87 so treat the y-axis as though it passes through the first point?
02:36.20 brlcad ideally first piece of the sketch (which might not be a point), but sure point would be a decent second best
02:36.38 brlcad first left-most point
02:37.02 poolio brlcad: sorry I've been so inactive the past week...hopefully I'll get more done this one :)
02:37.19 pacman87 i haven't actually used the sketch editor, so i dont' really know what im talking about
02:37.25 brlcad poolio: hope so :)
02:37.41 brlcad pacman87: you're not missing much :)
02:38.09 brlcad it was a *very* rudimentary interface to have some means to tweak sketches that were being imported from other CAD systems
02:38.17 pacman87 my parents and little bro are gone (college visiting), so i'm free to eat, sleep, and code for the next two days :)
02:38.37 brlcad not intended for day-to-day use really .. with sweeps and revolves using them, having a new sketch editor becomes a much higher priority ;)
02:38.45 brlcad maybe next gsoc follow-up ;)
02:42.56 brlcad pacman87: awesome (regarding intense coding) woot :)
02:43.34 brlcad pacman87: I hope you don't mind, but I sent out an announcement to our news mailing list about your progress on hyp
02:44.17 pacman87 that reminds me, i still need to finish those last few things
02:44.28 pacman87 revolve took over my brain too soon :)
02:48.09 pacman87 when's the june release going out?
02:50.32 brlcad don't worry about those few things, keep up on revolve
02:50.40 brlcad someone else can wrap up those other pieces
02:50.56 pacman87 someone else (tm)
02:51.01 brlcad document them into the TODO file
02:51.16 pacman87 ok, i wrote down the list
02:51.55 pacman87 it's a great feeling when i can find a mathematical proof for my intuition
02:52.05 brlcad hehe
02:58.30 CIA-22 BRL-CAD: 03brlcad * r31577 10/brlcad/trunk/src/conv/asc2g.c: good gravy, there were three different ws/indent styles going on in here. make it more consistent.
03:12.33 CIA-22 BRL-CAD: 03brlcad * r31578 10/brlcad/trunk/src/conv/asc2g.c: more cleanup. turn the comment blocks into doxygen blocks, reorder functions so forward declarations are not necessary.
03:14.58 *** join/#brlcad Axman6 (n=Axman6@pdpc/supporter/student/Axman6)
03:22.56 CIA-22 BRL-CAD: 03brlcad * r31579 10/brlcad/trunk/ (7 files in 5 dirs): deprecate HPT_LEN, HVECT_LEN, and ELEMENTS_PER_PT so that the defines are consistently ELEMENTS_PER_[type sans _t] across the board where they match the type they refer to.
03:27.37 poolio brlcad: do you think the nmg structure -> brep will be useful in the long wrong? For arb8 it seems simpler to just duplicate the tess() code
03:28.42 brlcad heh, subliminal slip? the long wrong sounds fun
03:29.56 poolio haha. long run. uh oh :)
03:30.15 brlcad no entiendo lo que dices
03:30.17 brlcad explain
03:31.10 poolio So, I'm working on just going from the struct model, the NMG model of the arb8 generated by tess and converting that to b-rep
03:31.50 poolio But it is going to be a pain in terms of keeping track of objects/etc... and it seems easier to just hard code values (similar to breplicator and such) and duplicate a lot of the code from tess()
03:32.04 CIA-22 BRL-CAD: 03brlcad * r31580 10/brlcad/trunk/TODO: woot, 3-fer-won .. the windows installer is better now, nsis goodness should be version aware now.
03:32.51 brlcad breplicator is a *horrible* example to follow for *anything* .. other than simply understanding how the ON_Brep type works
03:33.05 brlcad just so you know :)
03:33.09 pacman87 nice name ;)
03:33.31 brlcad there is/was no intention of keeping the tool around, it is a testing tool
03:33.50 brlcad same with the two brep_ examples
03:34.07 brlcad what objects do you have to keep track of?
03:34.16 brlcad for an arb, there's just one shell
03:34.36 poolio I understand that, I was just referencing the fact that to connect proper elements and such you need to keep track of them, and doing so with NMG seems more difficult than just doing it from a tess()-like method
03:35.11 brlcad quite possibly true
03:35.12 poolio Well, the vertices, edges, all the various elements
03:35.24 brlcad i'd see how many other tess() routines will be directly useful
03:35.27 poolio Although I suppose being able to work with the NMG may be useful in the future
03:35.52 brlcad say you wrote some completely generic struct model -> ON_Brep routine (which you're going to have to write anyways
03:36.05 poolio A lot of what I'm doing is kind of reverting the tess() work...it calculates what I need, puts it in NMG, and then I need to extract it. It'd be much easier to just work with the calculated values
03:36.17 brlcad e.g. to support NMG primitives.. :)
03:36.22 poolio fun fun :)
03:36.31 brlcad s/primitives/primitive/
03:36.59 brlcad not a huge deal, /me thinks -- your call
03:37.11 brlcad but I do think you'll end up needing that routine eventually
03:37.16 poolio does NMG support non-linear curves?
03:37.31 brlcad so if you have to tackle the hard problem, might as well avoid wasting the time redoing all the little easier problems
03:37.44 poolio Yeah, makes sense...if I write it properly than I'm done the NMG primitive work...maybe I should just start there and not work specifically on arb8
03:38.02 brlcad yeah, sorta -- the bspline primitive is built on top of nmg
03:38.27 brlcad (i.e. the old nurbs implementation)
03:38.47 poolio ah k. but NMG is just lines?
03:38.52 brlcad arbitrary bspline brep surfaces
03:39.13 poolio The paper I borrowed talks about future work to support more types of curves in NMG
03:39.20 brlcad nmg is technically just the structure (the 'uses') .. so it could be either
03:39.33 brlcad in practice, they're all linear segments afair
03:40.03 brlcad yeah, the entire bspline system is built on top of and/or into nmg
03:40.22 brlcad you have all the uses references, but the things they're using are geometry (which can be anything)
03:41.17 brlcad that's where opennurbs decided to pick a different set of terminology where historic used the 'uses' concept to distinguish topological structure from underlying geometry
03:42.15 CIA-22 BRL-CAD: 03brlcad * r31581 10/brlcad/trunk/TODO: I wrote a script that checks for missing files in a dist some time ago
03:43.32 *** part/#brlcad Byron1 (n=byron@pool-96-229-127-10.lsanca.fios.verizon.net)
03:43.49 CIA-22 BRL-CAD: 03brlcad * r31582 10/brlcad/trunk/TODO: promote testing the root solver patch
03:44.03 brlcad last Byron1
03:44.12 brlcad er, cya
03:51.52 CIA-22 BRL-CAD: 03brlcad * r31583 10/brlcad/trunk/TODO:
03:51.53 CIA-22 BRL-CAD: more cleanup, promote integrating last summer's opennurbs update and reviewing
03:51.53 CIA-22 BRL-CAD: the root solver. raytrace progress can be obtained by sending a signal
03:51.53 CIA-22 BRL-CAD: (SIGUSR1), adrt was updated/decoupled, and environ doesn't seem as interesting
03:51.53 CIA-22 BRL-CAD: as it used to.
04:00.13 CIA-22 BRL-CAD: 03brlcad * r31584 10/brlcad/trunk/TODO:
04:00.13 CIA-22 BRL-CAD: nice.. it's been hella long since I've apparently reviewed everything.. more
04:00.13 CIA-22 BRL-CAD: completed. indentation was run, eclectic was (briefly) reviewed, autogen.sh is
04:00.13 CIA-22 BRL-CAD: recursive, configure doesn't test stuff it doesn't need to, center of presented
04:00.13 CIA-22 BRL-CAD: areas are implemented, we have mirror around arbitrary vectors, and there's a
04:00.16 CIA-22 BRL-CAD: script to parse sf.net trackers
04:26.43 starseeker gets vmplayer working and is impressed by Haiku
04:26.47 CIA-22 BRL-CAD: 03brlcad * r31585 10/brlcad/trunk/ (27 files in 2 dirs): apparently pkg-config wants the files to have the same name as the internal name, who woulda thunkd it.
04:29.19 starseeker wonders how he's supposed to download stuff without a browser in the image...
04:29.51 starseeker ah ha - wget
04:29.56 starseeker of course
04:30.38 brlcad or curl
04:30.53 brlcad curl is slightly more prevalent than wget
04:31.20 starseeker huh. wonder how I stumbled into wget first
04:31.29 starseeker checks for curl...
04:32.41 brlcad i believe wget might be older and it's good for copying sites (recursively) .. curl is a generalized URL tool
04:32.50 starseeker ah
04:32.50 brlcad curl -O http://brlcad.org/index.html
04:33.13 brlcad curl http://brlcad.org/tmp/box.g > /tmp/box.g
04:33.51 starseeker ah, I'll bet that was it - recursive site copying
04:33.58 brlcad should find either on most linux distros
04:34.02 starseeker remembers attempting that with a few websites
04:34.09 brlcad bsd depends on what was installed
04:34.19 brlcad mac has curl by default, but no wget
04:34.36 brlcad windows has neither by default
04:34.51 brlcad forget what solaris did
04:35.51 starseeker is continually amazed by how little Windows has by default
04:38.53 starseeker you're mowing through stuff brlcad - nice work yourself ;-)
04:39.04 starseeker is reading scrollback...
04:39.28 brlcad not what I wanted to get to
04:39.59 starseeker What were you hoping for?
04:45.20 brlcad oh, didn't get to rtarea rewrite, libged/mged refactor bit, and new gui tested
04:48.31 starseeker ah
04:49.11 starseeker well, for what it's worth, nice image in the release news item :-) rtwizard I'm guessing?
04:49.27 brlcad hm?
04:49.36 brlcad nah, I just whipped that up on the fly
04:49.41 starseeker really
04:49.46 starseeker nice
04:49.55 starseeker 'course, I'm biased
04:50.46 brlcad :)
04:55.57 brlcad oo, there's a thought .. starseeker, have you tried tessellating a tire yet? with/without tread?
04:59.13 pacman87 hmmm, my numpad took over my mouse, and i'd don't know why (or how to turn it off)
04:59.21 pacman87 numlock doesnt' matter
05:02.46 starseeker nope
05:03.01 starseeker tried tessellating the mustang tire and it blew up
05:03.09 starseeker tries...
05:03.58 pacman87 starseeker's developing a new weapon
05:04.08 starseeker pheer the tire
05:04.19 pacman87 exploding tires
05:04.47 starseeker sees he has done one too many system updates since last recompiling BRL-CAD...
05:12.47 starseeker oo, nope - might have caught the libged transformation at a bad point... ok, gcc, do your stuff
05:20.31 starseeker wonders how long before 8 core supercomputers cost less than $300...
05:25.42 PrezKennedy you can get 4 core computers for not too bad
05:28.18 starseeker supposes he should just be more regular about recompiling...
05:31.19 starseeker let's try this...
05:31.58 starseeker facetize ftire tire
05:32.07 starseeker and the cpu goes nuts...
05:34.50 brlcad would expect it to take a long while
05:35.12 brlcad and .. um .. yeah, like at least daily :)
05:35.30 starseeker heh - I'll get going on the machine at work and forget to update my home machine
05:35.37 starseeker should set up a cron job...
05:35.54 brlcad I used to have it compile continuously via cron
05:36.14 starseeker wonders what brlcad's power bills were linke
05:36.18 starseeker er like
05:36.24 brlcad shrugs
05:36.30 brlcad i never turn my machines off
05:38.14 starseeker BRL-CAD - it's not just for modeling, it's also a space heating application!
05:39.31 starseeker well, considering this is the model WITHOUT tread, I'd say lord help anyone who has to do this
05:41.03 starseeker will let it run 'til morning
05:41.21 brlcad tries one with tread
05:41.51 starseeker can hear the machines flinching from here
05:44.37 brlcad wonders why OIS is trying to compile Linux interface files on Mac OS X
05:46.19 *** join/#brlcad Mouette (n=root@fw1.phys.sinica.edu.tw)
05:47.23 *** join/#brlcad clock_ (n=clock@77-56-95-212.dclient.hispeed.ch)
05:47.38 brlcad howdy Mouette
05:47.59 brlcad which I could stay longer, gotta run though .. hopefully the e-mails were enough info
05:50.16 Mouette ?
06:11.18 starseeker bemusedly wonders if ReactOS could run the windows BRL-CAD binary
07:04.30 *** join/#brlcad clock_ (n=clock@zux221-122-143.adsl.green.ch)
07:19.35 CIA-22 BRL-CAD: 03d_rossberg * r31586 10/brlcad/trunk/src/libged/CMakeLists.txt: duplicate entries removed
10:04.32 *** join/#brlcad Mouette (n=root@fw1.phys.sinica.edu.tw)
10:08.25 *** join/#brlcad elite01 (n=elite01@dslb-088-071-040-234.pools.arcor-ip.net)
11:00.35 *** join/#brlcad dtidrow (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net)
11:05.38 *** join/#brlcad mafm (n=mafm@elnet-111.lip.pt)
11:09.50 mafm hallo
11:14.54 *** join/#brlcad d_rossberg (n=rossberg@bz.bzflag.bz)
11:21.40 *** join/#brlcad thing0 (n=ric@123.208.49.174)
11:36.46 brlcad Mouette: I see why you were confused .. my messages hadn't sent, some bug in my e-mail client had them just sitting in my outbox
11:36.54 brlcad howdy mafm!
11:39.06 mafm hi brlcad
11:39.14 mafm did you see my messages in the log?
11:39.31 brlcad yep
11:39.57 mafm so did you try to build it? if so, any problems/comments?
11:40.20 brlcad I did, making progress, but it's quite far from seamless :)
11:40.38 brlcad not through getting the dependencies to build
11:41.34 mafm you mean the dependencies of OGRE?
11:41.41 brlcad no problems as yet, just annoyance of dependencies and each one having different build settings .. a couple are just crap build configurations on mac
11:41.45 mafm I think that the others don't have almost dependencies
11:42.10 mafm I see
11:42.27 brlcad rather linux-specific atm
11:43.33 mafm apart from that, I'd like to talk a bit about which functionalities create, how they should appear, etc
11:43.53 mafm in example, the taskbar of the video doesn't have "minimize"
11:44.14 mafm and I don't know if you'd like it or not,
11:44.57 mafm also I think that it would be good to have some window controls when clicking in the buttons of the taskbar (close, whatever)
11:45.32 mafm and having buttons in the taskbar somewhere for tiling/fullscreen, and things like that
11:49.58 mafm I have to go to lunch though, be back in a bit :)
11:49.58 brlcad minimization is an application modality
11:50.09 brlcad it handles it by providing new sheets
11:52.25 brlcad each tabbed sheet is effectively a new context, they're already effectively fullscreen unless there's a current 'task' in progress that might have caused a split or two
11:52.31 *** join/#brlcad SWPadnos (n=Me@dsl107.esjtvtli.sover.net)
11:53.02 brlcad from the *users* perspective, they shouldn't look or act like 'windows'
11:55.34 brlcad part of the point of non-overlapping contexts, and one of the big usability impacts is window management, not letting the user waste time with window layout/organization, overlaps, and active states (source of many modality errors)
11:56.24 brlcad that's talked about a fair bit throughout the IOE prototype
11:56.48 *** join/#brlcad Elperion (n=Bary@p5B14E41A.dip.t-dialin.net)
12:47.06 starseeker hrm. facetizing tire returned message: nmg_radial_build_list() min and max angle not adjacent in list (or list not monotone increasing)
12:47.24 *** join/#brlcad clock_ (n=clock@zux221-122-143.adsl.green.ch)
12:49.07 starseeker here's the full message: http://pastebin.bzflag.bz/m49a83cdf
12:50.03 *** join/#brlcad dtidrow_ (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net)
12:57.00 Mouette credit can be writen: Albatross Yang
12:57.40 Mouette you can rename it as BRL-CAD_7.12.4_solaris_x86.pkg
12:59.31 Mouette or other
13:00.23 Mouette but have keeped the version number and platform name.
13:09.42 ``Erik *snrkt* http://home.clara.net/lesmcdm/images/virgin.jpg (work safe)
13:10.01 ``Erik ish, anyways
13:13.04 brlcad that's *old* old :)
13:13.16 brlcad sees someone is making there way through /. comments
13:13.51 brlcad Mouette: can the two numbers that are inside the .pkg be manually edited safely?
13:14.12 brlcad the version numbers that are listed as 7.2.14 .. if they can, I'll just make the direct edits
13:16.19 Mouette you mean my version display 7.2.14 in my package?
13:16.27 brlcad yes
13:16.37 Mouette ok,i modify
13:16.45 brlcad er, maybe 7.4.12 .. one of the two
13:17.02 brlcad either way, it seemed .. not right :)
13:17.38 brlcad starseeker: interesting tessellation .. fails on the tread, but gets the inner tube and hub just fine
13:32.11 Mouette reuploading
13:44.52 PrezKennedy flexspending blows
13:45.11 *** join/#brlcad piotrnikiel (n=piotrnik@cy243.internetdsl.tpnet.pl)
13:55.49 mafm brlcad: the problem is that I was imagining modeling as something similar to using GIMP
13:56.11 mafm with many windows around to perform different operations on geometries
13:56.47 ``Erik (bsd has 'fetch' standard, curl and wget are "linux weenie tools")
13:58.06 ``Erik <-- *slow* slow, so it all works out
13:58.11 ``Erik and I'm caught up! w00t
13:58.52 Mouette i have done
14:01.29 brlcad Mouette: got it, looking it over
14:01.48 brlcad mafm: yes, that is a "problem" :)
14:02.00 ``Erik "margarita? I thought that was a slushy!"
14:02.44 brlcad mafm: that can be an option later (to allow broken out, overlapping window management), but isn't at all what we need first
14:03.00 mafm so I don't know which kind of contexts are necessary to create
14:03.52 ``Erik brlcad: do we have fastgen geometry stashed somewhere? there's some email noise about a fast4-g truncation error (ajem scr)
14:04.04 ``Erik (or is it fixed?)
14:04.15 brlcad two things to focus on, 1 - having one application window that can toggle between window and fullscreen and 2 - context/'window' management inside the application for the basic UI elements (taskbar, command overlay, graphics context, etc)
14:04.26 brlcad brlcad: see TODO
14:04.31 brlcad er ``Erik heh
14:04.40 ``Erik ah, heh
14:04.50 ``Erik doesn't answer my question, though, just the context statement
14:05.07 ``Erik well, answers my second question, but... :D
14:05.50 brlcad i have some somewhere, cad might on an unmounted disk
14:06.16 ``Erik heh, and no g-fast4.c :D
14:06.21 brlcad hence the todo, finding something to test was part of checking it out
14:06.31 brlcad yep, somewhat intentionally ..
14:06.54 brlcad there's not much point going from our format to theirs (from our perspective)
14:08.11 mafm brlcad: the application window is OGRE I guess, that it already works in that way (can be resized, and the gui windows that I create adapt to that)
14:08.40 brlcad i know it's ogre .. but can it go fullscreen? :)
14:08.42 mafm brlcad: then the command overlay it's already there, but without autocompletion or things like that
14:08.55 brlcad big window != fullscreen
14:09.13 mafm then the graphicx context would be the top panel I guess
14:09.17 prasad1 brlcad going fullscreen eh
14:09.23 mafm sure, it's a toggle :)
14:09.28 brlcad damn straight
14:09.47 mafm you can even start in fullscreen
14:09.59 mafm I just have to create a shortcut or something
14:10.24 prasad1 got a screenshot of current progress?
14:10.27 brlcad it'll probably eventually be a default
14:12.33 mafm the first time that you start, as usual OGRE apps, ask you for display configuration, and you can set it
14:12.50 mafm prasad1: http://wainu.ii.uned.es/~mafm/brlcad/brlcad_rbgui_20080623-1.png
14:14.51 brlcad mafm: you say you have the translucent command overlay working?
14:15.48 mafm not exactly... my initial idea of the console was a bit different, similar to IRC
14:15.54 brlcad when I mean IOE is a prototype, it's not just in basic functionality but in appearance too :)
14:16.01 brlcad we *also* need a console like you have
14:16.10 mafm hmmm
14:16.44 mafm so command overlay is the IOE-like one, and console IRC-like one, right?
14:16.52 brlcad right
14:16.58 mafm neat
14:17.27 brlcad so the console would be something that you could toggle on/off that would fit to a given side (e.g. to the bottom)
14:17.32 mafm the console one is also a full context?
14:17.50 mafm ah, good
14:18.12 mafm it's already working like that, only that I centered it thinking of the overlay
14:18.34 brlcad it's a panel similar to when in the IOE he does a search and it creates a results panel on the right
14:19.09 mafm so I think that I have enough work for a while, maybe then you or somebody else can test it and give more ideas
14:20.39 CIA-22 BRL-CAD: 03mafm * r31587 10/rt^3/trunk/src/g3d/ (GuiTaskbar.cxx GuiTaskbar.h): Making buttons of the taskbar to react when clicked
14:30.41 mafm brlcad: about shortcuts to enable different windows, fullscreen etc... are there guidelines, should I try to mimic something or should I invent them?
14:38.09 CIA-22 BRL-CAD: 03bob1961 * r31588 10/brlcad/trunk/ (7 files in 4 dirs): Added the l, listeval and paths commands to libtclcad. Split out the code for the l and paths commands (from wdb_obj.c) into separate files in libged and modified to use struct ged.
14:42.40 brlcad mafm: hook display of windows into the command window, some way to query what is available and enable/disable them
14:43.07 brlcad otherwise, mimic'ing IOE as close as possible in appearance and function should be more than enough work to keep you busy for a couple weeks
14:43.44 brlcad the only context missing is a graphics context, which is another packable widget type
14:43.54 brlcad i.e. something to display some 3D geometry in
14:44.01 prasad1 do these panels dock?
14:44.27 brlcad prasad1: eventually, they should be auto-docked by default
14:44.31 prasad1 i found the daz studio (free) interface pretty easy to use
14:44.35 prasad1 it's based on qt
14:44.36 brlcad prasad1: do you remember IOE?
14:44.51 prasad1 what does ioe mean
14:44.55 brlcad or had he not finished that by the time you left
14:45.07 brlcad it was an interaction prototype that jason worked on
14:45.25 prasad1 oh, prolly didnt start b4 i left
14:45.33 brlcad oh, he'd started
14:45.45 brlcad he'd started thinking about IOE before we left college :)
14:45.54 prasad1 oh heh
14:46.02 brlcad he just finally got to finishing a prototype for a class
14:46.28 prasad1 know which class?
14:46.28 brlcad it's pretty slick, lots of nice concepts
14:46.46 brlcad it was a usability analysis class iirc
14:47.01 prasad1 wait is this the thing he did for m3?
14:47.07 brlcad nope
14:47.08 prasad1 that got scratched
14:47.17 prasad1 *shrug*
14:47.40 brlcad it touches on some of the same concepts that m3 has/had, but really ends up being different
14:50.38 brlcad there's a video if you wanna watch
14:50.54 prasad1 utube?
14:51.00 brlcad uporn
14:51.08 prasad1 kinky
14:51.23 brlcad http://brlcad.org/design/gui/ioe_proto_final.mov
14:51.50 brlcad forget the task center and exact example he's going over, it's the ways you interact and core concepts that are the meat
14:52.01 brlcad takes about 10-15 min
14:54.06 prasad1 interesting
14:54.10 prasad1 something like quicksilver
14:54.12 brlcad mafm: so for starters, go for full-window tabs that you can toggle with the console being one and a graphics context being another .. with top task bar and bottom tabs
14:54.18 brlcad very much so
14:55.59 brlcad it's a cross between quicksilver and something like http://www.suckless.org/wiki/wmii and aqua usability concepts
14:58.50 mafm what do you mean bottom tabs -- the bottom tabs of MDI windows?
15:03.27 ``Erik O.o
15:04.52 prasad1 ioe is too much change for one political season
15:04.53 prasad1 ;)
15:05.46 brlcad mafm: no... the bottom tabs in ioe
15:05.50 CIA-22 BRL-CAD: 03mafm * r31589 10/rt^3/trunk/src/g3d/ (5 files): Creating Command Overlay and changing a bit the settings for the other windows
15:06.04 mafm mmm... the bottom tabs is what I call the taskbar
15:06.08 brlcad mdi tabs suck
15:06.21 mafm maybe you call the taskbar the upper panel? :D
15:07.16 brlcad there's the bar on top and tabbed selector on the bottom .. call them what you want
15:07.36 brlcad the one on the bottom should act like tabs, not buttons and not minimized windows
15:08.00 brlcad more like tabs in a browser
15:08.40 brlcad visually, that'st he one bit I didn't like -- it should be seamless to the context it activates (instead of appearing like a button)
15:09.10 mafm so, not restoring the appearance as a button, but having one highlighted at a given time
15:09.24 brlcad right
15:09.37 brlcad or rather, dehilighting the others, and having one just be "normal"
15:10.15 mafm I see
15:10.36 mafm I guess that I have to do it that way (or create the Tab myself, because they don't seem to have that in the library)
15:12.10 brlcad something sort of like this for the bottom: http://brlcad.org/tmp/tabs.png
15:12.26 brlcad where it blends up but demphasizes the others
15:14.18 mafm I see
15:14.59 CIA-22 BRL-CAD: 03starseeker * r31590 10/brlcad/trunk/doc/docbook/book/tutorial_series/VolumeII.xml: Fix docbook for authors so that it outputs cleanly
15:17.18 CIA-22 BRL-CAD: 03erikgreenwald * r31591 10/brlcad/trunk/src/libged/ (list.c pathsum.c): include missing header, cast to fix type warning (const issue)
15:19.25 mafm I guess that I should create some kind of "window manager" class to register the windows, the specialized unique windows (taskbar, etc)
15:19.56 mafm and to take high-level layout decisions, such as tiling windows
15:24.57 brlcad absolutely, that was part of the reason why I pointed you at wmii .. to see if maybe there was even some layout code that could be pulled from it
15:25.46 CIA-22 BRL-CAD: 03brlcad * r31592 10/brlcad/trunk/misc/pkgconfig/Makefile.am: libwdb typo
15:26.10 brlcad they have two basic layout options, one being tabbed, the other being 'stacked'/shaded
15:28.01 mafm I wonder if I can use wmii with KDE :D
15:37.41 CIA-22 BRL-CAD: 03starseeker * r31593 10/brlcad/trunk/doc/docbook/ (18 files in 2 dirs): Point all the lessons and the book itself to a single authors file in the tutorial series directory, since they're all pulled from the book
15:40.28 mafm huh
15:40.40 mafm not with KDE, but it's working now :D
15:41.55 CIA-22 BRL-CAD: 03bob1961 * r31594 10/brlcad/trunk/src/libged/pathsum.c: Changed TCL_ERROR to GED_ERROR in a few places.
16:17.51 CIA-22 BRL-CAD: 03bob1961 * r31595 10/brlcad/trunk/ (5 files in 5 dirs): Changed a handfull of function name prefixes from bu_ bu_tcl_.
16:29.27 CIA-22 BRL-CAD: 03homovulgaris * r31596 10/brlcad/trunk/ (8 files in 4 dirs): Stage 1/4 towards a simple binary constraint solver: Addition of simple Network and Solution class
16:38.10 *** join/#brlcad Elperion (n=Bary@p5B14E41A.dip.t-dialin.net)
16:42.48 pacman87 hmmm, so i have an arbitrary number of hitpoints to store and sort in shot(), but using a linked list requires memory allocation during shot. extrude just assumes there will be no more that 64 hitpoints
17:06.35 CIA-22 BRL-CAD: 03mafm * r31597 10/rt^3/trunk/src/g3d/ (11 files):
17:06.35 CIA-22 BRL-CAD: Introducing Window Manager class, which will perform high-level operations. One
17:06.35 CIA-22 BRL-CAD: example of the future functionality is that it took over the window resized
17:06.35 CIA-22 BRL-CAD: events: instead of registering all individual windows for render window resized
17:06.35 CIA-22 BRL-CAD: events, only the window manager gets notified and then it delivers the event to
17:06.39 CIA-22 BRL-CAD: some of the windows.
17:30.55 CIA-22 BRL-CAD: 03mafm * r31598 10/rt^3/trunk/src/g3d/ (5 files):
17:30.55 CIA-22 BRL-CAD: Removing Taskbar as separate window, and reintroduce it instead as a 'frame'
17:30.55 CIA-22 BRL-CAD: window, part of the WindowManager class. This makes sense since WindowManager
17:30.55 CIA-22 BRL-CAD: is already going to have the list of available windows, it should subtract the
17:30.56 CIA-22 BRL-CAD: size of the taskbar when passing the available screen size to the rest of the
17:30.58 CIA-22 BRL-CAD: windows, etc.
17:34.07 CIA-22 BRL-CAD: 03pacman87 * r31599 10/brlcad/trunk/ (7 files in 3 dirs): beginning of revolve primitive, with shot() algorithm for straigh line sketches (untested)
17:48.15 mafm uh
17:48.24 mafm fullscreen works, but not very smoothly for me :D
17:50.02 mafm well, when adding more resolution it works much better really
17:53.45 CIA-22 BRL-CAD: 03mafm * r31600 10/rt^3/trunk/src/g3d/ (Application.cxx Application.h): Adding fullscreen support
17:59.57 starseeker Here's the html conversion as it currently stands: http://my.bzflag.bz/~starseeker/vol2/book/tutorial_series/
18:00.12 starseeker that stupid unicode character is at is again...
18:00.28 starseeker why doesn't that happen on the local copy????
18:32.53 CIA-22 BRL-CAD: 03mafm * r31601 10/rt^3/trunk/src/g3d/ (10 files):
18:32.53 CIA-22 BRL-CAD: Adding a bar in the top of the screen, that will be used for context actions,
18:32.53 CIA-22 BRL-CAD: set fullscreen and similar tasks. Application class was made a Singleton so it
18:32.53 CIA-22 BRL-CAD: can be accessed from elsewhere, in example to toggle fullscreen mode from the
18:32.54 CIA-22 BRL-CAD: recently created top bar.
18:48.32 starseeker Ah hah, getting closer - http://my.bzflag.bz/~starseeker/vol2/book/tutorial_series/
18:50.31 mafm ~starseeker++
19:21.36 CIA-22 BRL-CAD: 03starseeker * r31602 10/brlcad/trunk/doc/docbook/README: Add toplevel docbook readme with a few notes
19:26.15 *** join/#brlcad dtidrow (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net)
19:56.41 mafm have to go now, take care guys
19:56.45 pacman87 bye mafm
19:57.27 mafm http://wainu.ii.uned.es/~mafm/brlcad/brlcad_rbgui_20080624-1.png
19:57.28 mafm :)
20:45.23 CIA-22 BRL-CAD: 03bob1961 * r31603 10/brlcad/trunk/ (6 files in 4 dirs): Added the attr command to libtclcad. Split out the code for the attr command (from wdb_obj.c) into a separate file in libged and modified it to use struct ged.
20:50.52 *** join/#brlcad louipc (n=louipc@76-10-146-181.dsl.teksavvy.com)
20:52.14 *** join/#brlcad thing0 (n=ric@123.208.16.63)
21:36.55 *** join/#brlcad the_drizzzle (n=matt_s@124-197-9-145.callplus.net.nz)
21:37.44 *** join/#brlcad the_drizzle (n=matt_s@124-197-9-145.callplus.net.nz)
22:13.50 brlcad pacman87: you could preallocate a pool of memory during prep, put that into the specific structure, then only allocate if you run out
22:14.13 pacman87 best of both worlds strategy?
22:14.19 brlcad can use bu_vls or bu_vlb as automatic pools
22:14.28 brlcad (probably vlb)
22:14.47 brlcad yeah, it's a decent practical balance
22:18.35 brlcad starseeker: I had them reversed, the server declaration does indeed override the meta declaration so the .htaccess and/or a section in httpd.conf that declares specific subtrees as utf-8 would be needed
22:20.43 brlcad default setting for all sites is ISO-8859-1, so that makes more sense now
22:21.25 brlcad could possibly change the whole server to utf-8, but would need a weekend to verify it doesn't break any sites
23:11.43 *** join/#brlcad dtidrow_ (n=dtidrow@c-69-255-182-248.hsd1.va.comcast.net)
23:34.29 CIA-22 BRL-CAD: 03brlcad * r31604 10/brlcad/trunk/TODO: implement make, ted, edsol, mirror, and analyze for hyp primitive
23:35.02 brlcad pacman87: does that magic number mean anything or did you pick random chars?
23:35.17 pacman87 i think it's 'hype'
23:35.27 pacman87 might be caps
23:35.38 brlcad should document it in the header :)
23:35.45 brlcad same for revolve
23:39.08 pacman87 in magic.h?
23:40.26 brlcad yep
23:40.39 pacman87 ok
23:58.20 CIA-22 BRL-CAD: 03brlcad * r31605 10/brlcad/trunk/src/libpc/ (pcNetwork.h solver_test.cpp): start class name in caps

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