IRC log for #brlcad on 20100216

00:35.34 Ralith kanzure: swigged to python?
00:36.03 kanzure perhaps :)
02:06.11 ``Erik huh, didn't know those has a vertical launch position... http://www.my914-6.com/crap/
03:38.34 brlcad ``Erik: heh, yeah, I'd seen that
03:42.28 brlcad kanzure: depends what exactly you're talking about
03:43.40 brlcad in terms of a swigified libged interface, I'd start with the direct translation of mged's command API
03:44.04 brlcad that's a lot more like python's file_handler interface
03:45.23 brlcad you'd do something like ged_obj = GED.open("file.g", "r") .. ged_obj.make("sphere", "sph") .. ged_obj.close()
03:46.48 brlcad so you wouldn't yet need to worry about geometry object classes, you wouldn't have geometry objects, everything would go through the defined ged interface like is done procedurally
03:46.58 brlcad that is a direct / familiar one-to-one mapping
03:48.32 brlcad once the Geometry Engine comes closer to completion (which is at least a year out), that would be the way to swigify geometry objects themselves with a one-to-one API mapping, but not with libged
03:48.37 brlcad baby steps :)
03:49.40 brlcad kanzure: think about how you might turn this into a python script: http://brlcad.org/wiki/SGI_Cube
03:50.06 brlcad that's a *much* easier first-step swig mapping
03:53.41 kanzure brlcad: i'm not too happy about a procedural ged wrapper with things like ged_obj.make("primitive name", "abbrev") :/
03:53.55 kanzure the csg.py file i showed was just my attempt at figuring out what i would be comfortable with
03:54.47 kanzure if you refresh csg.py ( http://designfiles.org/~bryan/csg.py ) you'll see a small update
03:54.53 kanzure i'd be very happy with that :)
04:18.17 brlcad kanzure: it really depends what the goal is -- what you're showing in your snippet is not easily achieved
04:18.34 brlcad and doesn't directly mirror a completed API
04:18.58 brlcad the closest would be the Geometry Service, which is a C++ interface that has similar object mappings and methods
04:20.01 brlcad i'd be happy with what you're showing, I just don't think it's realistic without implementing a large portion of the Geometry Service in python directly
04:20.06 kanzure it wasn't meant to be brlcad specific :(
04:20.07 brlcad certianly not swigable
04:20.18 kanzure i was thinking of making this a layer on top of pythonOCC
04:20.28 brlcad what is python occ?
04:20.30 kanzure http://designfiles.org/skdb/geom/geom.py
04:20.40 kanzure http://pythonocc.org/ it's a swig interface to OpenCASCADE (OCC)
04:20.51 brlcad ah
04:20.59 kanzure it's what i've been using for the past few months
04:22.01 brlcad then your comments are only more confusing :)
04:22.11 brlcad because you started saying you want to swigify brlcad
04:23.21 brlcad if you're looking to add CSG to an opencascade interface.... I'm not sure I really care then, to be honest :)
06:37.52 *** join/#brlcad cjdevlin (~devlin@99-74-181-148.lightspeed.cicril.sbcglobal.net)
08:49.54 *** join/#brlcad cjdevlin (~devlin@99-74-181-148.lightspeed.cicril.sbcglobal.net)
10:58.13 *** join/#brlcad Phurl (~mdupont@ip-81-210-228-126.unitymediagroup.de)
11:09.43 *** join/#brlcad Phurl_ (~mdupont@ip-81-210-228-126.unitymediagroup.de)
13:05.21 *** join/#brlcad CoconutCrab (~toor@210.86.231.65)
13:46.52 *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1)
13:56.35 *** join/#brlcad CoconutCrab (~toor@210.86.231.65)
13:57.04 *** join/#brlcad CoconutCrab (~toor@unaffiliated/coconutcrab)
14:34.07 *** join/#brlcad parigaudi (~quassel@pd95b7f5e.dip0.t-ipconnect.de)
15:26.05 CIA-43 BRL-CAD: 03brlcad * r37629 10/brlcad/trunk/HACKING:
15:26.05 CIA-43 BRL-CAD: once again, the FRS instructions have changed. there is now a virtualization
15:26.05 CIA-43 BRL-CAD: service that you start up which gives you a 'normal' shell with access to the
15:26.05 CIA-43 BRL-CAD: various project services. from there, the frs files can be accessed and
15:26.06 CIA-43 BRL-CAD: managed.
15:27.38 CIA-43 BRL-CAD: 03brlcad * r37630 10/brlcad/trunk/HACKING: need quotes, otherwise ambiguous target
15:28.04 ``Erik ah, 'file release system'
15:31.07 CIA-43 BRL-CAD: 03brlcad * r37631 10/brlcad/trunk/HACKING: add a note to point out that this can all be done through the web interface too.
15:35.12 ``Erik verde... wow, nifty O.o and from a highschool project no less
15:43.22 CIA-43 BRL-CAD: 03brlcad * r37632 10/brlcad/trunk/HACKING: minor tweak, don't need to call exit since we didn't pass -t in order to get a tty
15:50.00 CIA-43 BRL-CAD: 03bob1961 * r37633 10/brlcad/trunk/src/ (51 files in 15 dirs): More updates to silence warnings on 64-bit windows.
16:32.23 CIA-43 BRL-CAD: 03brlcad * r37634 10/brlcad/trunk/HACKING: release steps to reflect improved openness. always notify our three main release channels (main website, project site, mailing list) and freshmeat when releases are made, regardless of how major/minor they are.
16:32.42 CIA-43 BRL-CAD: 03starseeker * r37635 10/brlcad/branches/dmtogl/ (278 files in 71 dirs): Update dmtogl branch to r37633
16:37.05 CIA-43 BRL-CAD: 03bob1961 * r37636 10/brlcad/trunk/src/ (3 files in 3 dirs): Update the CAD version.
17:04.37 *** join/#brlcad Computer_ (~Computer@209-16-114-100.net.bhntampa.com)
18:14.18 starseeker brlcad: what would be involved with adding support for dbip->dbi_base2local type information for mass, time, etc?
18:15.07 starseeker would like to have mass, time, etc. have the same global defaults as length, if possible
18:33.48 starseeker e.g:mged> units
18:33.49 starseeker Length: Millimeters (mm) (1 mm = 1 mm)
18:33.49 starseeker Mass: Kilograms (kg) (1 kg = 1000 g)
18:33.49 starseeker Time: Seconds (s) (1 s = 1 s)
18:35.47 starseeker I looks like it would involve modifying the db_i struct, which I'm guessing is trouble...
19:48.22 CIA-43 BRL-CAD: 03starseeker * r37637 10/brlcad/trunk/ (BUGS NEWS src/rt/viewweight.c): Fix bug reported by Dwayne Kregel where incorrect units were being reported by rtweight
20:54.55 kanzure where can i find a definition of a sphere?
20:54.56 kanzure http://brlcad.org/xref/source/src/libged/
20:59.57 kanzure ah maybe http://brlcad.org/xref/source/src/librt/primitives/
21:09.53 CIA-43 BRL-CAD: 03erikgreenwald * r37638 10/brlcad/trunk/src/adrt/load_g.c: make BoT fast-loading work.
21:18.13 ``Erik sweet, loading a bot converted m35 in under a second
21:26.54 starseeker aweeet
21:26.58 starseeker sweet even
21:27.42 ``Erik (dang pointers)
22:16.33 *** join/#brlcad QAChip (~Christian@201.122.75.250)
22:17.03 QAChip Hello all, Im having problems to compile BRL, can anyone help me
22:24.35 yukonbob QAChip: post a spefic question and you're more likely to get a reply...
22:24.42 yukonbob *specific
22:27.23 *** join/#brlcad QAChip (~Christian@201.122.75.250)
22:28.00 QAChip <PROTECTED>
22:28.39 QAChip #1.1, if not, I'll paste my outputs in fedora's pastebin
22:31.36 QAChip http://www.fpaste.org/vG6Z/
22:32.30 starseeker uh... - what's the problem? type make
22:32.46 QAChip This is the output of ./configure, it says it will also make several packages that I already have installed, like, Tcl, Tk, ITcl/ITk, etc
22:33.08 QAChip the problem is that ./configure is not recognizing my installed packages
22:33.10 starseeker it may be that the versions you have installed won't work for BRL-CAD
22:33.20 starseeker it' shouldn't matter
22:33.47 starseeker it will install local copies for BRL-CAD
22:33.54 QAChip ok, i have another issue when typing make, I'll send you the output... hold on
22:38.17 QAChip ...making...
22:43.03 QAChip ...still making...
22:49.51 QAChip ..still there...
22:57.04 QAChip hahah, no problem now, perhaps last update fixed the issue
22:58.27 QAChip I gotta go, I'll be back to tell you what happened, anyway, thx, see you all later
23:57.09 ``Erik dang turds knocked my laptop down to the floor

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