| 01:06.05 | ``Erik | the national cathedral is quite an impressive structure, and not just cuz it has darth vader carved into it O.o |
| 01:52.05 | *** join/#brlcad IriX64 (~MarioDUli@bas2-sudbury98-1177872136.dsl.bell.ca) |
| 02:23.27 | CIA-43 | BRL-CAD: 03starseeker * r40019 10/brlcad/branches/cmake/ (CMakeLists.txt misc/CMake/FindGL.cmake): |
| 02:23.27 | CIA-43 | BRL-CAD: By default, CMake's opengl logic assumes AGL is the opengl of interest on Apple, |
| 02:23.27 | CIA-43 | BRL-CAD: and requires manual specification for anything else. We'll need to do better |
| 02:23.27 | CIA-43 | BRL-CAD: than that - add visus file as an example of how to approach this, although it's |
| 02:23.27 | CIA-43 | BRL-CAD: not likely to be the final solution. |
| 02:30.38 | *** join/#brlcad yukonbob (~svs@S010600235a187d92.ok.shawcable.net) |
| 02:34.07 | brlcad | starseeker: mpfr is used under the hod by gmp, which would be a lot "easier" to apply to our code |
| 02:34.34 | brlcad | have to make a couple things work first, though |
| 02:35.13 | brlcad | entire build with c++ compiler (CC=g++) and consistent fastf_t use |
| 02:53.13 | starseeker | erm |
| 02:53.26 | starseeker | has never tried CC=g++... |
| 02:55.35 | starseeker | heh |
| 02:55.36 | starseeker | checking for gcc... (cached) g++ |
| 02:55.37 | starseeker | checking whether the C compiler works... no |
| 02:55.47 | starseeker | okay then |
| 03:08.44 | yukonbob | starseeker: sometimes as simple as a non-executable /tmp |
| 03:11.05 | CIA-43 | BRL-CAD: 03starseeker * r40020 10/brlcad/branches/cmake/CMakeLists.txt: whoops, do the right malloc check. |
| 04:34.46 | brlcad | starseeker: I've cleaned up specific libraries to compile in c++ mode in the past |
| 04:35.03 | brlcad | it's actually a pretty good sanity/portability test |
| 05:14.43 | brlcad | starseeker: curious that yours failed on the sanity check, it should get past it unless other "default" flags are in play (that are incompatible) |
| 05:14.59 | brlcad | it will fail in the code, though, anyways |
| 05:15.24 | brlcad | one thing you can't get away with at all is k&r-isms, which we still have some of, even in our headers |
| 05:16.11 | brlcad | functions the same name as structs are also not allowed (conflicts with their constructor) |
| 05:16.26 | brlcad | nothing hard, but it would be a week or two of cleanup |
| 05:17.03 | brlcad | was waiting for this minor to fix at least some of the issues like the constructor conflicts |
| 06:55.07 | CIA-43 | BRL-CAD: 03d_rossberg * r40021 10/rt^3/tags/rel-7-16-10/: tag the C++ core interface with the corresponding BRL-CAD version (i.e. 7.16.10) |
| 07:05.52 | *** join/#brlcad merzo (~merzo@smartbussiness.mobicom.net.ua) |
| 10:04.07 | *** join/#brlcad mafm (~mafm@231.Red-80-26-128.dynamicIP.rima-tde.net) |
| 11:04.47 | *** join/#brlcad Stattrav (~Stattrav@117.192.128.37) |
| 14:12.33 | *** join/#brlcad Zaebos (~irc@pd95b7f5e.dip0.t-ipconnect.de) |
| 14:14.05 | *** join/#brlcad Alexandrus (~nil@p4FE3F3B1.dip.t-dialin.net) |
| 14:14.09 | Alexandrus | moin(); |
| 14:46.34 | *** join/#brlcad packrat (~packrator@99-67-225-40.lightspeed.livnmi.sbcglobal.net) |
| 14:54.06 | brlcad | hello Alexandrus |
| 14:54.11 | Alexandrus | hello brlcad |
| 14:54.16 | Alexandrus | i have a simple question |
| 14:54.33 | Alexandrus | how do you modell a tube with sqare ends |
| 14:54.38 | Alexandrus | (flat ends) |
| 14:55.39 | brlcad | that'd be two overlapping arb8's making a + then cylinders down the sides to round the edges |
| 14:56.20 | Alexandrus | so a great number of cylinders |
| 14:56.30 | Alexandrus | thought maybe there is a tgc mod hidden somewhere |
| 14:56.33 | brlcad | well, great number == 4 |
| 14:56.38 | Alexandrus | 4? |
| 14:56.41 | brlcad | 4 |
| 14:56.49 | brlcad | 2 arb8's and 4 rccs |
| 14:57.20 | Alexandrus | ok, the 2 arb8's for the ends |
| 14:57.25 | Alexandrus | but the 4 rccs??? |
| 14:57.30 | Alexandrus | one for the tube |
| 14:57.32 | Alexandrus | and 3? |
| 14:57.37 | brlcad | no no |
| 15:00.01 | Alexandrus | http://kccm.dnsalias.org:8080/sketch1.png |
| 15:00.02 | brlcad | so consider the shape in 2d first |
| 15:00.20 | brlcad | two *overlapping* rectangles, then circles on the corners |
| 15:00.27 | brlcad | like this http://www.jeremydjohnson.com/images/post_images/2008/11/box_sections.gif |
| 15:01.20 | brlcad | you diagram is just a slightly more complicated for where you have multiple bends, but the exact same idea applies |
| 15:01.58 | Alexandrus | the problem is the smooth interval between section 1-2 and 2-3 |
| 15:02.13 | Alexandrus | i don't see how i do this with rcc |
| 15:02.27 | Alexandrus | or tor.. |
| 15:03.16 | brlcad | imagine that in that image, sections 1+2+3 ARE the first arb8 rectangle |
| 15:03.17 | Alexandrus | guess maybe i need an example |
| 15:03.54 | brlcad | then there's another overlapping arb8 crossing through the other two horizontal sides and section 2 |
| 15:03.58 | brlcad | those are unioned together |
| 15:04.14 | Alexandrus | not seeing the use of a second arb8, but ok.. |
| 15:04.21 | Alexandrus | now we have a "cross" |
| 15:04.25 | brlcad | a "+" |
| 15:05.34 | Alexandrus | ah, i see what you mean... |
| 15:06.08 | Alexandrus | guess this doesn't work with constant surface though |
| 15:06.25 | Alexandrus | (seen from end to end, surface must stay constant) |
| 15:06.40 | brlcad | what do you mean by constant? |
| 15:06.59 | Alexandrus | very pratical, you get a tube and flatten it (hammer, heat) |
| 15:07.24 | Alexandrus | A(x)=const |
| 15:07.29 | Alexandrus | (approx) |
| 15:08.38 | brlcad | ah, a taper |
| 15:08.49 | Alexandrus | otherwise i could just + the tube with two recs.. |
| 15:08.56 | Alexandrus | hmm... |
| 15:09.04 | brlcad | you can make that work, that's just a tgc with different ends |
| 15:09.26 | brlcad | rcc == tgc with equal ends, tgc is generalized where they're not equal |
| 15:09.36 | Alexandrus | hmm |
| 15:09.44 | Alexandrus | and then plus |
| 15:10.06 | Alexandrus | ok, this might work.. |
| 15:11.41 | Alexandrus | still...i think 1 arb8, one rcc and two tgc's should do the job |
| 15:11.50 | Alexandrus | unless i have forgotten something |
| 15:12.10 | Alexandrus | yes...the cut...another rcc and two tgc |
| 15:14.07 | brlcad | so sort of like this: http://www.cemframework.com/img/CrusaderSolid.png |
| 15:14.33 | Alexandrus | hmm...just much more soft |
| 15:14.47 | Alexandrus | i could try using two outer tgc at one end |
| 15:15.05 | Alexandrus | to get a smooth curve from rcc to arb8 |
| 15:15.10 | brlcad | yeah, the cylinder would taper and there's another arb8 that would run tangent at 90 degrees, then a big arb8 in the middle to fill it in |
| 15:15.44 | Alexandrus | the second arb8 isn't necessary to what i can see now |
| 15:16.17 | Alexandrus | i will try modelling it |
| 15:16.56 | brlcad | for a 4-sided rounded edge tube, you will have to use at least 3 arb8's and 4 tgc's |
| 15:17.37 | brlcad | or 5 arb8's and 4 tgcs for a more optimally oranized tube with minimally overlapping primitives |
| 15:18.10 | brlcad | at least in order to capture an arbitrary taper |
| 15:18.13 | brlcad | depends on the shape |
| 15:18.29 | Alexandrus | i guess i will just modell one end first |
| 15:18.31 | Alexandrus | show it to you |
| 15:18.38 | Alexandrus | and you give me advice on how to improve it |
| 15:18.59 | Alexandrus | the second end is equal, so shouldn't be a problem |
| 15:42.40 | CIA-43 | BRL-CAD: 03starseeker * r40022 10/brlcad/branches/cmake/ (3 files in 2 dirs): Ah, handy - Ogre has FindCarbon and FindCocoa files - add them in. There may be a few other items of interest in Ogre's CMake packages. |
| 15:53.54 | Alexandrus | @brlcad: http://kccm.dnsalias.org:8080/render1.png |
| 15:54.02 | Alexandrus | this is a rough approximation |
| 16:02.25 | *** join/#brlcad Stattrav (~Stattrav@117.192.152.56) |
| 16:06.18 | Alexandrus | i have to go shopping..see you later |
| 17:17.11 | *** join/#brlcad Alexandrus (~nil@p4FE3F3B1.dip.t-dialin.net) |
| 17:17.19 | Alexandrus | re |
| 19:00.30 | CIA-43 | BRL-CAD: 03starseeker * r40023 10/brlcad/branches/cmake/CMakeLists.txt: Expand the tests for functions and headers. It looks like some of this should probably be handled by Find* files, but still figuring out how to approach this. |
| 19:07.20 | *** join/#brlcad IriX64 (~MarioDUli@bas2-sudbury98-1128565464.dsl.bell.ca) |
| 19:27.34 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) |
| 19:34.29 | *** part/#brlcad cjdevlin (~devlin@99-74-181-148.lightspeed.cicril.sbcglobal.net) |
| 19:44.15 | CIA-43 | BRL-CAD: 03starseeker * r40024 10/brlcad/branches/cmake/misc/CMake/FindPkgMacros.cmake: Whoops - OGRE macros are using routines in this file. Probably will need to do some utility routine consolidation later. |
| 19:58.55 | *** join/#brlcad IriX64 (~MarioDUli@bas2-sudbury98-1128565320.dsl.bell.ca) |
| 20:05.22 | CIA-43 | BRL-CAD: 03starseeker * r40025 10/brlcad/branches/cmake/ (3 files in 3 dirs): Tweak things a bit - for some reason generating the brlcad_config.h in the out-of-dir build location isn't working for including, need to investigate. |
| 20:18.26 | starseeker | hah, this is actually quite interesting: http://booksearch.blogspot.com/2010/08/books-of-world-stand-up-and-be-counted.html |
| 20:29.53 | CIA-43 | BRL-CAD: 03starseeker * r40026 10/brlcad/branches/cmake/CMakeLists.txt: |
| 20:29.53 | CIA-43 | BRL-CAD: Gonna need to set the PNG options ahead of the ADD_SUBDIRECTORY call in the |
| 20:29.53 | CIA-43 | BRL-CAD: THIRD_PARTY macro unless we want to require two passes before building - has its |
| 20:29.53 | CIA-43 | BRL-CAD: drawbacks since if an external png lib is used these options are useless, but |
| 20:29.53 | CIA-43 | BRL-CAD: mark as advanced to hide them from casual site. Could be worse - it also |
| 20:29.54 | CIA-43 | BRL-CAD: highlights specific functionality needed by BRL-CAD. |
| 21:07.02 | CIA-43 | BRL-CAD: 03n_reed * r40027 10/brlcad/trunk/src/tclscripts/ (3 files in 2 dirs): working button commands in bot editor plus minor interface changes |
| 21:27.28 | starseeker | bemusedly notes that none of the sourceforge binary files for linux or bsd seem to follow the convention... |
| 21:29.27 | starseeker | yipe - binary file is almost 400 megs... |
| 21:29.31 | starseeker | checks why... |
| 21:33.23 | starseeker | ooh, right, docbook html files, lots of new libraries since last binaries... (wow it has been a long time...) |
| 21:35.54 | *** join/#brlcad cjdevlin (~devlin@99-74-181-148.lightspeed.cicril.sbcglobal.net) |
| 22:03.32 | starseeker | winces - this is gonna be one doozie of an upload... |
