| 00:16.47 | ``Erik | heh, yeah, uh, the bottie merge there took a few hours |
| 00:17.51 | ``Erik | I imagine I'll have to do a quick followup tomorrow along with the "number_of_triangles" hack and backing the ray up a micron, then see how bad I screw trunk up |
| 00:20.54 | starseeker | hrm |
| 00:21.31 | starseeker | possible issue... |
| 00:21.52 | starseeker | libbn/mat.c:54: error: MAT_INIT_IDN undeclared here (not in a function) |
| 00:26.15 | starseeker | tries a clean build |
| 00:29.37 | ``Erik | did any conflicts list? |
| 00:29.52 | starseeker | not in CMake (at least not that I saw) |
| 00:29.58 | ``Erik | during the merge |
| 00:30.10 | starseeker | yeah, didn't see any |
| 00:30.31 | ``Erik | funky, I always seem to get at least one |
| 00:31.00 | starseeker | I do get some on occasion, usually when I've done something in the cmake branch first and then in trunk |
| 00:31.11 | starseeker | yeah, libbn builds now |
| 00:31.34 | starseeker | oh, bet it was using the old copy in the build directory's include dir |
| 00:31.40 | starseeker | dur |
| 00:32.47 | CIA-58 | BRL-CAD: 03starseeker * r42610 10/brlcad/branches/cmake/ (249 files in 76 dirs): Update cmake branch to trunk r42609 |
| 00:34.12 | CIA-58 | BRL-CAD: 03starseeker * r42611 10/brlcad/branches/cmake/ (include/CMakeLists.txt src/mged/CMakeLists.txt): Remove files no longer present from CMakeLists.txt files |
| 00:43.00 | starseeker | brlcad: is that new rtuif.h header supposed to be installed? |
| 00:43.39 | ``Erik | it's the new rt_private.h, I don't think it is |
| 00:43.53 | starseeker | k |
| 00:44.08 | starseeker | ah, right |
| 00:44.14 | starseeker | in src/rt, not include |
| 00:45.53 | brlcad | it's a private header, fixed a problem |
| 00:46.18 | starseeker | nods - yeah, was reading the list wrong, my bad |
| 01:00.33 | starseeker | interesting I'm getting a bu_ipwd crash with the cmake build on mac |
| 01:02.41 | brlcad | hm, entirely possible -- I didn't get the chance to test my fix before heading in to the meeting |
| 01:02.56 | brlcad | any particular test sequence? |
| 01:03.01 | starseeker | Program received signal EXC_BAD_ACCESS, Could not access memory. |
| 01:03.01 | starseeker | Reason: KERN_PROTECTION_FAILURE at address: 0x00000000 |
| 01:03.01 | starseeker | 0x93a266b7 in realpath$DARWIN_EXTSN () |
| 01:03.12 | brlcad | realpath was given a null |
| 01:03.17 | starseeker | libbu/brlcad_path.c:100 |
| 01:03.34 | starseeker | I don't see how, and gdb doesn't think ipwd is null... |
| 01:04.09 | brlcad | hm |
| 01:04.18 | brlcad | might be a difference between bsd and linux realpath() |
| 01:04.27 | brlcad | what does your manual page say for the second argument? |
| 01:05.09 | brlcad | I pass NULL for the result since bsd realpath() will use that as a key to malloc and pass the result back as the return value |
| 01:05.19 | brlcad | the first param shouldn't be null.. |
| 01:05.39 | brlcad | oh, DARWIN .. you're on mac too.. |
| 01:05.53 | starseeker | ah, that could be why - I think the proper handling of second argument being NULL in realpath is only guaranteed by POSIX 2008 |
| 01:08.31 | starseeker | considered snarfing GNU's canonicalize_file_name function and putting that into libbu, but it kinda seemed like overkill... |
| 01:10.37 | CIA-58 | BRL-CAD: 03brlcad * r42612 10/brlcad/trunk/include/bu.h: check for PATH_MAX before _MAX_PATH |
| 01:13.17 | CIA-58 | BRL-CAD: 03starseeker * r42613 10/brlcad/branches/cmake/src/libbu/brlcad_path.c: realpath + NULL is not happy on Mac, apparently - will this work? |
| 01:17.39 | starseeker | brlcad: that seems to work here, if you don't mind the static solution - IIRC realpath only runs into trouble on operating systems like HURD that have unlimited path length, and somehow I doubt we'll be worried about running on HURD anytime soon... |
| 01:17.54 | starseeker | heads out |
| 01:19.40 | CIA-58 | BRL-CAD: 03brlcad * r42614 10/brlcad/trunk/src/libbu/brlcad_path.c: unlike 10.6 and recent linux, realpath() on Mac OS X 10.5 is not set up to take NULL for the second paramter so pass in our path buffer and adjust logic accordingly. |
| 01:20.50 | brlcad | similar solution |
| 01:24.20 | brlcad | usually best to avoid function call return values as expressions unless the function specifically returns a boolean (e.g., isspace()); unintentional pointers and integers are common logic bugs |
| 01:59.22 | CIA-58 | BRL-CAD: 03brlcad * r42615 10/brlcad/trunk/TODO: promote plate mode nurbs since it's scheduled for Q2, create a new project section for NURBS with added tasks for implementing implicit CSG to NURBS CSG, documenting the new primitive, and boolean evaluation. |
| 02:15.23 | CIA-58 | BRL-CAD: 03brlcad * r42616 10/brlcad/trunk/TODO: expand, consolidate and clean up the section for NMG/BoT too |
| 02:43.53 | *** join/#brlcad yukonbob (~bch@S0106001cf044d085.ok.shawcable.net) | |
| 02:43.54 | yukonbob | oh hai |
| 03:08.31 | DX^ | Anyone alive that could explain to me how NURBS are rendered |
| 03:08.40 | DX^ | does the program break it up into triangles? |
| 03:08.52 | DX^ | I don't understand how it could otherwise show smooth surfaces |
| 03:20.48 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) | |
| 03:58.11 | starseeker | DX^: it doesn't break it up into triangles |
| 03:58.40 | starseeker | it builds a surface tree, which provides an initial guess for a hit point which is then iterated to a solution |
| 03:59.15 | starseeker | we'll get triangles eventually for tessellation and shaded displays, but the are only an approximation - the NURBS surface itself is more accurate |
| 04:21.38 | starseeker | brlcad: doesn't that patch return the success/fail value of realpath instead of the result in buffer? |
| 04:24.11 | starseeker | oh, nevermind - I see |
| 04:29.31 | starseeker | nifty - realpath returns what you need as the result. not bad |
| 04:30.53 | CIA-58 | BRL-CAD: 03starseeker * r42617 10/brlcad/branches/cmake/ (5 files in 4 dirs): Sync to trunk 42616 (mostly, brlcad_path.c patch is approximate since the Windows version of realpath is unaddressed in trunk). |
| 05:16.25 | CIA-58 | BRL-CAD: 03brlcad * r42618 10/brlcad/trunk/TODO: little section for STEP-related tasks |
| 08:16.33 | *** join/#brlcad Yoshi47 (~jan@64.235.102.210) | |
| 09:13.26 | *** join/#brlcad Stattrav (~Stattrav@117.192.131.143) | |
| 09:13.37 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) | |
| 10:48.06 | *** join/#brlcad epileg (~epileg@unaffiliated/epileg) | |
| 12:18.21 | starseeker | glares at falling snow and unplowed roads |
| 12:31.08 | starseeker | hmm... slashdot redesigned again |
| 12:35.01 | starseeker | gets introduced to the Stylish Firefox plugin... nifty |
| 13:12.24 | CIA-58 | BRL-CAD: 03starseeker * r42619 10/brlcad/trunk/src/tclscripts/ (archer/Archer.tcl mged/man.tcl): Tweak man page viewer to work in-build-dir |
| 13:47.07 | brlcad | <PROTECTED> |
| 13:47.28 | brlcad | I'd think the embedded path separators will cause problems on windows |
| 13:47.36 | brlcad | that's what file join is supposed to resolve |
| 13:47.42 | DaveLo | Mernin all |
| 13:47.53 | brlcad | howdy DaveLo |
| 13:48.12 | DaveLo | Hows you? take the test yet? |
| 13:49.40 | brlcad | not yet, still have another week or two of studying |
| 13:49.55 | DaveLo | ah, okie. |
| 13:50.02 | CIA-58 | BRL-CAD: 03davidloman * r42620 10/rt^3/trunk/docs/ (4 files): Adding in some .psd's for the wiki graphics. |
| 13:50.24 | DaveLo | stares at the falling snow and is eager to get out there later and make the Ultimate Snow Fortress |
| 14:01.48 | starseeker | brlcad: ok, I can try that |
| 14:06.11 | CIA-58 | BRL-CAD: 03starseeker * r42621 10/brlcad/trunk/src/tclscripts/ (archer/Archer.tcl mged/man.tcl): Don't assume '/' for dir separators |
| 14:08.39 | starseeker | ah, the mged one was just a typo (not sure why it didn't show up before...) archer wasn't using file join where it needed to |
| 14:09.29 | CIA-58 | BRL-CAD: 03starseeker * r42622 10/brlcad/branches/cmake/ (5 files in 4 dirs): Update cmake branch to trunk r42621 |
| 14:39.20 | *** join/#brlcad Stattrav (~Stattrav@117.192.146.62) | |
| 14:39.20 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) | |
| 14:41.48 | DaveLo | ``Erik: why is TCL needed in rt3? |
| 14:45.33 | brlcad | DaveLo: tcl.h is included by raytrace.h and bu.h |
| 14:46.09 | brlcad | hoping we can eliminate that inclusion at some point in the future, but decoupling from tcl in the public API is a job in itself |
| 14:46.40 | DaveLo | ah, i see. |
| 14:47.29 | DaveLo | is the tcl libsj included in the windows brlcad binaries install? |
| 14:47.49 | brlcad | should be |
| 14:47.57 | brlcad | won't run without em |
| 14:48.09 | DaveLo | thats what I thought. |
| 14:48.29 | DaveLo | for somereason, i was building rt3 just fine, but now that there is a findTCL.cmake in the works, its failing. |
| 14:48.33 | DaveLo | :/ |
| 14:49.57 | brlcad | might have been pulling some system tcl.h |
| 14:50.25 | DaveLo | not likely, as i don't have tcl.h on my windows system ;) |
| 14:50.35 | brlcad | hm, that's curious then |
| 14:51.12 | brlcad | not only why it's now needing it, but how it worked before too |
| 14:51.34 | DaveLo | i know cmake can find the brlcad include and libs, i just don't think the cmake find is smart enough to dive down in to a subdir of brlcad/include |
| 14:52.02 | DaveLo | I think it was working before because if cmake can find bu.h, then bu.h knows where the tcl headers |
| 14:52.05 | DaveLo | are |
| 14:52.28 | DaveLo | cmake, however, doesn't use bu.h to gain a clue where the tcl headers are. |
| 14:52.29 | DaveLo | :/ |
| 14:54.42 | DaveLo | bah, that's too deep of a rabit hole to dive in atm :/, i'll leave windows till later i guess. |
| 14:56.13 | brlcad | bu.h doesn't know, it just does a #include "bu.h" |
| 14:56.16 | brlcad | er, tcl.h |
| 14:58.07 | brlcad | so something else, something really basic, is going on |
| 14:59.30 | DaveLo | likely something else. :/ |
| 15:01.47 | brlcad | fwiw, include paths are build system trivialities, any dev should be able to diagnose/understand/override fix with ease |
| 15:02.39 | brlcad | otherwise it's like a builder not knowing how a power drill works :) |
| 15:03.48 | DaveLo | heh, well in that case, Im still learning about the power drills, so don't make fun =P |
| 15:04.31 | brlcad | not making fun, just saying "don't fear the power drill" |
| 15:04.51 | brlcad | or treat it like it's untouchable black magic that will take a long time to figure out |
| 15:04.55 | ``Erik | dlo: tcl is required for bu.h |
| 15:05.16 | ``Erik | when using a system that doesn't have tcl.h in /usr/include or /usr/brlcad/include, it was failing |
| 15:07.01 | ``Erik | *readreadread* heh, yeh |
| 15:07.08 | brlcad | cmake adds a little extra complexity, but it's akin to adding a pneumatic compressor to your drill .. it's still a drill though ;) |
| 15:07.19 | DaveLo | well wait... if tcl.h is installed into a non standard location, it was failing? |
| 15:07.25 | ``Erik | probably need a -DTCL_INCLUDE_DIR=C:\some\path\to\brlcad\include |
| 15:07.48 | DaveLo | either of you two going to be in on Friday? |
| 15:07.57 | ``Erik | the problem was that it was failing when tcl was installed to the standard include path... not the path that linux and BRL-CAD expect |
| 15:08.22 | DaveLo | what 'standard include path' are you talking about/ |
| 15:08.24 | DaveLo | ? |
| 15:08.30 | ``Erik | I have no plans to not be in on friday... depends on weather |
| 15:08.33 | ``Erik | /usr/local/ |
| 15:09.43 | ``Erik | mac and linux both abuse the system header directory for user packages like package managed tcl |
| 15:14.11 | ``Erik | O.O gtk3? hmmm |
| 15:35.18 | DaveLo | configure --enable-all or configure --enable-everything |
| 15:35.21 | DaveLo | i cannot remember |
| 15:50.48 | brlcad | they're aliases for the same flag if it was either of those |
| 17:27.39 | DaveLo | 5" on the ground so far. Second wave is coming in a few hours! |
| 17:45.25 | *** join/#brlcad Stattrav (~Stattrav@117.192.143.147) | |
| 17:45.25 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) | |
| 18:26.48 | DaveLo | Okay, round one is shoveled. Now, bring on round two! |
| 18:50.13 | CIA-58 | BRL-CAD: 03Dloman 07http://brlcad.org * r2446 10/wiki/GSNet_String: |
| 18:55.03 | CIA-58 | BRL-CAD: 03Dloman 07http://brlcad.org * r2447 10/wiki/GSNet_String: Update table styling |
| 18:56.10 | *** join/#brlcad Yoshi47 (~jan@64.235.102.210) | |
| 18:57.43 | CIA-58 | BRL-CAD: 03Dloman 07http://brlcad.org * r2448 10/wiki/GSNet_String: Adjust table width |
| 19:02.00 | CIA-58 | BRL-CAD: 03Dloman 07http://brlcad.org * r2449 10/wiki/GeometryServiceNetworkProtocol: Fixed headline |
| 19:30.12 | CIA-58 | BRL-CAD: 03Dloman 07http://brlcad.org * r2450 10/wiki/NetMsgTypes: headline fixes and spacing. |
| 19:49.33 | *** join/#brlcad Stattrav (~Stattrav@117.192.143.147) | |
| 19:49.33 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) | |
| 20:07.41 | *** join/#brlcad CIA-53 (~CIA@208.69.182.149) | |
| 20:15.24 | CIA-53 | BRL-CAD: 03brlcad * r42623 10/brlcad/trunk/src/libged/ls.c: eliminate dead code |
| 20:18.57 | *** join/#brlcad Ralith (~ralith@d142-058-078-138.wireless.sfu.ca) | |
| 20:22.52 | CIA-53 | BRL-CAD: 03brlcad * r42624 10/brlcad/trunk/src/librt/db_lookup.c: calloc our ptbl so we're initialized to zero |
| 20:34.59 | CIA-53 | BRL-CAD: 03brlcad * r42625 10/brlcad/trunk/src/librt/db_lookup.c: v4 geometry database do not use d_major_type and d_minor_type, so zero-set accordingly as a preventive measure. |
| 20:39.47 | brlcad | https://github.com/mcneel/rhinocommon |
| 20:40.04 | brlcad | (they just released another sdk as open source) |
| 20:41.37 | brlcad | it's the C# portion of Rhino 5.0's new cross-platform SDK |
| 20:47.06 | CIA-53 | BRL-CAD: 03brlcad * r42626 10/brlcad/trunk/src/conv/asc/g2asc.c: do not check values directly. _GLOBAL is merely an attribute-only object. |
| 22:17.18 | *** join/#brlcad Ralith (~ralith@142.58.92.37) | |
| 22:20.43 | *** join/#brlcad yukonbob_ (~bch@20-144.wireless.kamloops.net) | |
| 22:20.49 | yukonbob_ | hello, #brlcasd |
| 22:20.56 | yukonbob_ | #brlcad, even. |
| 22:25.04 | yukonbob_ | png.h has 2 diff't prototypes for PNG_EXPORT(). |
| 22:25.44 | yukonbob_ | or... 1s; /me reviews |
| 22:29.20 | *** join/#brlcad R0b0t1 (~Enigma@64-136-219-55.dyn.everestkc.net) | |
| 22:29.20 | *** join/#brlcad R0b0t1 (~Enigma@unaffiliated/r0b0t1) | |
| 22:38.13 | yukonbob_ | :P nn --- conflicting header files from diff't versions of libpng. |
| 22:38.25 | yukonbob_ | *nm |
| 22:51.02 | yukonbob_ | q: how far along in the build process is one if asc2g conversions are running? |
| 23:16.10 | brlcad | http://itmanagement.earthweb.com/osrc/article.php/3922041/50-Open-Source-Applications-for-Sci-Tech-Education.htm |
| 23:16.22 | brlcad | yukonbob_: basically the build is done |
| 23:16.39 | brlcad | all of the sources have compiled by that point |
| 23:23.08 | yukonbob_ | brlcad: congratulate me, then. I've got brlcad built on netbsd for first time in ~2 years. |
| 23:23.28 | yukonbob_ | then fails, but I think I know why. I'll have some patches forthcoming. |
| 23:25.44 | CIA-53 | BRL-CAD: 03brlcad * r42627 10/brlcad/trunk/TODO: |
| 23:25.44 | CIA-53 | BRL-CAD: pull up annotation primitive since it's soon. also relaly need to upgrade the |
| 23:25.44 | CIA-53 | BRL-CAD: back-end svn repo so branches can be tracked better (especially since there is a |
| 23:25.44 | CIA-53 | BRL-CAD: major merge coming soon). push down dbcp/merg and exporting all top-level |
| 23:25.44 | CIA-53 | BRL-CAD: objects in the converters. |
| 23:25.44 | brlcad | excellent |
| 23:25.53 | brlcad | it's been a while since I've done a netbsd build myself |
| 23:27.07 | yukonbob_ | I've been working on this now for last ~24 hours (not non-stop :) ) and think I've got the roadblocks solved for this particlar case... which is NetBSD -current, Tcl8.6b, and some other oddities. |
| 23:28.53 | yukonbob_ | runtime will be another test itself... I'm not intimately familiar w/ itcl/itk, and it's a major-version bump over what brl-cad ships with... so that's be interesting. |
| 23:32.26 | CIA-53 | BRL-CAD: 03brlcad * r42628 10/brlcad/trunk/src/libged/ls.c: |
| 23:32.26 | CIA-53 | BRL-CAD: fix a bug with the ls command where it was printing NULL or garbage for the |
| 23:32.26 | CIA-53 | BRL-CAD: object type if you requested an ls -la long listing. this only occurred for v4 |
| 23:32.26 | CIA-53 | BRL-CAD: geometry databases but was due to an assumption in the code that d_minor_type is |
| 23:32.26 | CIA-53 | BRL-CAD: a suitable index into the rt_functab[] (which it's not). other commands need |
| 23:32.26 | CIA-53 | BRL-CAD: fixing too. |
| 23:44.02 | CIA-53 | BRL-CAD: 03brlcad * r42629 10/brlcad/trunk/src/libged/attr.c: attributes aren't valid for v4, but don't use d_minor_type as an index into rt_functab[] regardless. |
| 23:53.55 | yukonbob_ | default installation == /usr/local/*? |
| 23:58.09 | yukonbob_ | sees is /usr/brlcad |