| 01:39.28 | CIA-48 | BRL-CAD: 03starseeker * r41734 10/brlcad/branches/cmake/ (4 files in 3 dirs): |
| 01:39.28 | CIA-48 | BRL-CAD: Getting closer on Windows. Irritatingly, the dlls are being put in bin not lib |
| 01:39.29 | CIA-48 | BRL-CAD: even after adding the extra config redircts (need to generalize that along these |
| 01:39.29 | CIA-48 | BRL-CAD: lines: http://forum.openscenegraph.org/viewtopic.php?t=6926) so all of the |
| 01:39.30 | CIA-48 | BRL-CAD: pkgIndex.tcl scripts being generated will have to be customized to look for |
| 01:39.30 | CIA-48 | BRL-CAD: their dlls in ../../bin on Windows." |
| 02:16.08 | *** join/#brlcad crazy_imp (~mj@a89-182-219-17.net-htp.de) | |
| 03:27.25 | *** join/#brlcad cjdevlin (~devlin@d118-75-252-178.try.wideopenwest.com) | |
| 06:38.04 | *** join/#brlcad dtidrow (~dtidrow@c-68-60-96-218.hsd1.mi.comcast.net) | |
| 06:50.48 | *** join/#brlcad merzo (~merzo@227-89-94-178.pool.ukrtel.net) | |
| 10:33.44 | *** join/#brlcad cjdevlin (~devlin@d118-75-252-178.try.wideopenwest.com) | |
| 10:55.20 | *** join/#brlcad mafm (~mafm@170.Red-88-18-69.staticIP.rima-tde.net) | |
| 12:30.12 | *** join/#brlcad mafm (~mafm@170.Red-88-18-69.staticIP.rima-tde.net) | |
| 12:30.49 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) | |
| 13:05.53 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) | |
| 13:58.37 | *** join/#brlcad Zaebos (~irc@pd95b7f5e.dip0.t-ipconnect.de) | |
| 14:25.53 | CIA-48 | BRL-CAD: 03starseeker * r41735 10/brlcad/branches/cmake/CMakeLists.txt: Generalize the output directory settings along the lines hinted at by http://forum.openscenegraph.org/viewtopic.php?t=6926 |
| 14:27.11 | CIA-48 | BRL-CAD: 03starseeker * r41736 10/brlcad/branches/cmake/CMakeLists.txt: Whoops, typo. |
| 14:49.40 | CIA-48 | BRL-CAD: 03starseeker * r41737 10/brlcad/branches/cmake/src/other/incrTcl/ (itcl/CMakeLists.txt itk/CMakeLists.txt): Start with itcl/itk - try tweaking the build directory pkgIndex.tcl to use CMAKE_LIBRARY_OUTPUT_DIRECTORY. |
| 15:51.44 | brlcad | starseeker: curious, how'd you get all of the c99 stdint.h types to work on windows without pstdint.h? |
| 15:54.34 | starseeker | I didn't - but pstdint.h was already being included in brlcad_config.h |
| 15:54.56 | starseeker | so having it in config_win AND brlcad_config was resulting in a double-include, which VC++ didn't like |
| 15:56.19 | CIA-48 | BRL-CAD: 03starseeker * r41738 10/brlcad/branches/cmake/src/other/incrTcl/ (itcl/CMakeLists.txt itk/CMakeLists.txt): Sigh - we need the lib prefix on Windows, and aren't getting it - try this. |
| 16:04.46 | brlcad | aaah! that makes complete sense then :) |
| 16:09.19 | starseeker | I'm actually plesantly surprised - I expected a lot more to blow up when I started including brlcad_config.h on Windows, but so far (knock on wood) it's been fairly painless |
| 16:55.07 | brlcad | awesome! |
| 18:05.25 | CIA-48 | BRL-CAD: 03starseeker * r41739 10/brlcad/branches/cmake/src/other/tcl/library/CMakeLists.txt: Fix some version numbers on .tm library file installs |
| 18:20.27 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) | |
| 18:34.36 | *** join/#brlcad mafm (~mafm@244.Red-88-23-77.staticIP.rima-tde.net) | |
| 18:34.37 | CIA-48 | BRL-CAD: 03starseeker * r41740 10/brlcad/branches/cmake/ (include/optical.h src/libmultispectral/init.c): Tweaks to liboptical and libmultispectral to get multispectral building on Windows |
| 18:45.36 | starseeker | rtsrv defines it's own bu_log and bu_bomb - that's causing compile issues on Windows |
| 18:46.56 | starseeker | ah - looks like remrt and rtsrv haven't been ported to windows |
| 18:57.40 | cjdevlin | i came across this site: mil-oss.org and didn't find brl-cad in the existing projects list http://mil-oss.org/get-involved/existing-projects - just wondering if you guys were familiar w/ this? |
| 19:07.32 | DaveLo | brlcad: check this out: http://www.revolvermaps.com/ |
| 19:07.49 | DaveLo | might be cool for the brlcad.org site :) |
| 19:30.32 | CIA-48 | BRL-CAD: 03davidloman * r41741 10/rt^3/trunk/include/NetMsgTypes.h: WS, Formatting. |
| 19:33.56 | brlcad | cjdevlin: yes, I spoke and gave a presentation on BRL-CAD at Mil-OSS WG2 just a couple months ago -- just haven't bothered anyone to get added to that list yet |
| 19:34.49 | brlcad | http://mil-oss.org/about/news-and-announcements/6-events/29-mil-oss-wg2 |
| 19:53.03 | CIA-48 | BRL-CAD: 03starseeker * r41742 10/brlcad/branches/cmake/ (3 files in 3 dirs): Get ssamp-bw building on Windows |
| 19:57.03 | brlcad | cjdevlin: the initial list was just directly pulled from a report |
| 20:02.14 | cjdevlin | brlcad: ahh |
| 20:02.34 | cjdevlin | brlcad: i thought grass was also an army corps of engineers project |
| 20:03.05 | CIA-48 | BRL-CAD: 03starseeker * r41743 10/brlcad/branches/cmake/src/rt/ext.h: Try including optical.h here. |
| 20:16.00 | brlcad | cjdevlin: it was |
| 20:16.28 | brlcad | though they didn't really release it as open source, they were basically done with the project so they put it into the public domain |
| 20:17.45 | brlcad | researchers in italy (iirc) and elsewhere picked up the PD code, made substantial sustained contributions and as a community made a new claim of copyright enabling them to license under GPL |
| 20:19.34 | CIA-48 | BRL-CAD: 03starseeker * r41744 10/brlcad/branches/cmake/src/CMakeLists.txt: Bah - comment out some stuff that clearly is not Windows ported - deal with this later. |
| 20:20.18 | cjdevlin | brlcad: last i checked their web site still claimed nasa and some other government organizations as users. it seems like these 2 projects (brlcad and grass) are two of the exemplary models of open source in the military models. |
| 20:21.14 | cjdevlin | brlcad: it seems like a lot of other projects on that list were developed outside of the military and are now used by the military whereas these 2 were created in house and now have stable code bases with at least a few outside contributors |
| 20:21.31 | brlcad | starseeker: yeah, and they way they're define bu_bomb/bu_log, they override the calls made to those functions *within* our own libraries |
| 20:22.07 | brlcad | cjdevlin: there are a couple others now, but brl-cad and grass are definitely exemplary for the u.s. army |
| 20:22.41 | brlcad | i have a public google docs that lists a lot of other significant DoD and Gov open source projects |
| 20:22.57 | brlcad | can't get to it at the moment, but it might be searchable |
| 20:23.18 | brlcad | *I* can't get to it at the moment, you can if you can find it :) |
| 20:24.16 | cjdevlin | brlcad: i will try and take a look. thanks for the info. |
| 20:24.30 | starseeker | brlcad: wouldn't that come under the heading of "Bad Things"? Or was that a deliberate design decision? |
| 20:24.55 | brlcad | starseeker: funky tricks with C |
| 20:24.58 | brlcad | deliberate |
| 20:25.10 | starseeker | ugh. Windows Not Happy with it |
| 20:26.37 | brlcad | yeah, other platforms too -- usually takes some trickery per new platform |
| 20:32.09 | brlcad | reworking them to not override the library would be good |
| 20:44.46 | CIA-48 | BRL-CAD: 03starseeker * r41745 10/brlcad/branches/cmake/ (include/raytrace.h src/rt/rtshot.c): Try adding rt_shootray_bundle and rt_raybundle_maker to raytrace.h - will see if this works on Windows. |
| 20:53.34 | CIA-48 | BRL-CAD: 03starseeker * r41746 10/brlcad/branches/cmake/ (include/raytrace.h src/rt/rtshot.c): Hmm - probably better to assume these were not added to raytrace.h for a reason, see if the same version in rtshot.c will suffice. |
| 21:06.56 | brlcad | putting them into the public header effectively "publishes" those API calls, meaning we'd be reasonably expected to follow deprecation steps should we want to change/remove them later |
| 21:07.05 | brlcad | not putting them in the public header means that they can be changed at will |
| 21:07.37 | brlcad | they're experimental functions .. now probably obsolete with the new rt_shootrays() interface |
| 21:17.36 | starseeker | On Windows it can't resolve them |
| 21:23.34 | CIA-48 | BRL-CAD: 03starseeker * r41747 10/brlcad/branches/cmake/src/librt/ (bundle.c mkbundle.c): OK, cheat - we must export these for rtshot on Windows, but they're not part of the public API, so put the export stuff in the C file next to the functions, to be removed when they are. |
| 21:25.37 | brlcad | starseeker: it can't resolve them due to declarations and definitions |
| 21:25.56 | brlcad | that has nothing to do with them being in the public header or not |
| 21:25.57 | starseeker | nods - I got it with the last commit, builds now |
| 21:51.01 | CIA-48 | BRL-CAD: 03r_weiss * r41748 10/brlcad/trunk/src/libbn/bntester.c: |
| 21:51.02 | CIA-48 | BRL-CAD: Updated the 'bntester' tool used to test 'libbn' functions. The basic framework |
| 21:51.03 | CIA-48 | BRL-CAD: is complete and has been setup to test the 'bn_distsq_line3_pt3' function (i.e. |
| 21:51.04 | CIA-48 | BRL-CAD: function number 1). Additional 'libbn' functions can be added as necessary. |
| 21:51.14 | CIA-48 | BRL-CAD: 03brlcad * r41749 10/brlcad/trunk/TODO: |
| 21:51.15 | CIA-48 | BRL-CAD: more details on the tgc bug. it's somehow only calculating one hit for rays |
| 21:51.15 | CIA-48 | BRL-CAD: that perfectly graze the edge of the cylinder's lengthwise tube surface (not the |
| 21:51.16 | CIA-48 | BRL-CAD: end disks or actual topological edges). bug may be specific to cones that pinch |
| 21:51.16 | CIA-48 | BRL-CAD: to a point (like s.nos5g) but unconfirmed. |
| 21:55.56 | starseeker | Down to these errors (plus a few that disappear on recompile and I suspect are dependencies I haven't yet specified fully): http://paste.lisp.org/display/117939 |
| 22:06.03 | brlcad | pixcmp is probably just not defining the DLL import flag |
| 22:07.31 | brlcad | hurt.c is exactly a reduced all-in-one-file version of the rt source files, so whatever fix is there that makes rt work on windows should make the hurt sources work |
| 22:08.17 | brlcad | dunncomm is deprecated, can ignore |
| 22:08.54 | brlcad | csgbrep is accessing a bunch of functions directly and should not -- should be going through the functab interface |
| 22:12.05 | brlcad | tie.h defines a variable named 'far' which is a reserved word for msvc, so needs to be renamed |
| 22:12.37 | brlcad | the rest is mostly fluff |
| 22:12.43 | brlcad | easy fixes |
| 22:13.23 | brlcad | pixblend.c needs to use the libbu timer and random number facilities |
| 23:00.27 | *** topic/#brlcad by brlcad -> BRL-CAD Open Source Solid Modeling || http://brlcad.org || http://sf.net/projects/brlcad || #brlcad logs: http://ibot.rikers.org/%23brlcad/ || BRL-CAD release 7.18.0 is posted (20101209) || Happy Open Source Anniversary 2010-12-21 !!! Six years... | |
| 23:04.53 | starseeker | brlcad: what's a good example of going through the functab? |
| 23:05.16 | starseeker | (csgbrep was just a test utility anyway, it's probably not critical to have it on Windows) |