| 00:02.30 | bhinesley | starseeker: cursor.c errors are back http://paste.pocoo.org/show/458445/ |
| 00:24.03 | bhinesley | starseeker: scratch that, appears to be my fault |
| 00:28.44 | abhi2011 | ok I get an error in analyze.c : /home/abhi/socis/brlcad/src/libged/analyze.c:312:5: error: call to function ârt_arb_centroidâ without a real prototype |
| 00:29.00 | bhinesley | starseeker: double scratch that... not my fault |
| 00:29.44 | abhi2011 | some more : /home/abhi/socis/brlcad/include/raytrace.h:3354:23: note: ârt_arb_centroidâ was declared here |
| 00:30.17 | abhi2011 | seems the declaration is not being taken correctly |
| 00:31.02 | abhi2011 | bhinseley: your error seems also related to implicit declarations |
| 00:31.57 | bhinesley | nods |
| 00:32.52 | abhi2011 | hmm even in track.c its the same thing: error: call to function âtrack_mk_combâ without a real prototype |
| 01:25.35 | CIA-62 | BRL-CAD: 03bhinesley * r45979 10/brlcad/trunk/src/libged/edit.c: |
| 01:25.37 | CIA-62 | BRL-CAD: During certain batch translations, objects after the first target object simply |
| 01:25.37 | CIA-62 | BRL-CAD: moved to the same location as the first target object. This was due to certain |
| 01:25.37 | CIA-62 | BRL-CAD: translation vectors not being recalculated after executing the first |
| 01:25.37 | CIA-62 | BRL-CAD: translation. Since information about an argument is lost after the first vector |
| 01:25.37 | CIA-62 | BRL-CAD: is calculated, there needs to be a new vector (and therefore struct edit_arg) |
| 01:25.38 | CIA-62 | BRL-CAD: for each translation. |
| 01:53.12 | starseeker | bhinesley: can you give me a make VERBOSE=1 with the cursor error? |
| 01:53.55 | starseeker | also, can you tell me if brlcad_config.h has #define HAVE_TERMCAP_H 1 ? |
| 01:57.13 | bhinesley | there is no #define HAVE_TERMCAP_H |
| 01:57.19 | starseeker | growl |
| 01:57.19 | bhinesley | I'm running make right now |
| 01:57.33 | starseeker | one sec... let me see if I accidentally undid my fix |
| 01:57.48 | bhinesley | I saw it there |
| 02:01.13 | CIA-62 | BRL-CAD: 03starseeker * r45980 10/brlcad/trunk/src/other/CMakeLists.txt: Oops - variables changed, so update conditionals that use them... |
| 02:01.18 | starseeker | bhinesley: give that a shot |
| 02:01.39 | starseeker | sigh... the hazards of radical rework |
| 02:04.05 | bhinesley | rebuilding |
| 02:07.10 | dli | starseeker, does cmake scripts support system tcl/tk now? |
| 02:07.28 | dli | starseeker, instead of brlcad local building |
| 02:09.02 | starseeker | dli: it should - testing that now |
| 02:09.28 | CIA-62 | BRL-CAD: 03starseeker * r45981 10/brlcad/trunk/src/other/CMakeLists.txt: Be slightly more aggressive with the find_library command for itcl |
| 02:09.56 | dli | starseeker, great, because gentoo people still don't accept my packages, due to mandatory tcl/tk local building |
| 02:10.09 | starseeker | winces - yeah, not surprised |
| 02:13.54 | starseeker | dli: still working on local itcl/itk - one second |
| 02:21.23 | *** join/#brlcad juanman (~quassel@unaffiliated/juanman) |
| 02:34.08 | CIA-62 | BRL-CAD: 03starseeker * r45982 10/brlcad/trunk/src/ (3 files in 3 dirs): |
| 02:34.09 | CIA-62 | BRL-CAD: Ah, right. Original CMake logic written assuming package require based Itcl/Itk |
| 02:34.09 | CIA-62 | BRL-CAD: usage - when C libs were again necessary, just hacked in quickly. Can't get away |
| 02:34.09 | CIA-62 | BRL-CAD: with that anymore, so use variables where appropriate and look for itk C library |
| 02:34.15 | starseeker | dli: give that a shot |
| 02:36.02 | dli | starseeker, revision 45982 |
| 02:36.16 | dli | starseeker, testing now |
| 02:36.23 | starseeker | that works for me here with system tcl/tk and itcl/itk - I don't have a system iwidgets, tkhtml, tkpng or tktable |
| 02:36.53 | starseeker | the raster toolkit, opennurbs and STEP Class Libraries will stay on because we're currently maintaining our own versions of those |
| 02:38.02 | starseeker | dli: also, remember the cmake options have changed a lot - still probably in flux, actually |
| 02:40.35 | dli | starseeker, this is reported configure options: |
| 02:40.38 | dli | starseeker, http://pastebin.com/g7dD2VMP |
| 02:41.21 | starseeker | hmm - surprized the optimized release summary option isn't printing |
| 02:41.36 | starseeker | is that a clean configure from a fully updated trunk? |
| 02:41.53 | starseeker | what configure options are you passing in? |
| 02:42.37 | bhinesley | starseeker: builds fine, thanks |
| 02:43.40 | dli | starseeker, options passed to cmake: http://pastebin.com/h3ELaNfD |
| 02:44.16 | starseeker | dli: ah, let me translate those |
| 02:44.58 | starseeker | dli: will gentoo accept a BRL-CAD build of tkhtml? |
| 02:46.08 | dli | starseeker, should be fine, since there's no tkhtml package in gentoo yet |
| 02:46.50 | starseeker | they don't want static libs? |
| 02:48.36 | dli | starseeker, I think it's ok, but they want to be sure about how libs are used/linked, so, package manager can avoid broken libraries better, package deps better |
| 02:50.34 | starseeker | dli: um - surprised you turned off the example geometry - is that a policy? |
| 02:51.04 | starseeker | with EXTRADOCS off, the html documentation won't be available |
| 02:51.56 | dli | starseeker, it's controlled by USE flag: examples |
| 02:52.43 | dli | starseeker, http://pastebin.com/7ndb9WDQ |
| 02:53.16 | dli | starseeker, so, USE="doc" will enable extradocs, extradocs_pdf, extradocs_man |
| 02:54.20 | dli | starseeker, aqua, does brlcad support aqua? |
| 02:54.35 | starseeker | no |
| 02:55.41 | dli | starseeker, then, it's a mistake |
| 02:58.02 | starseeker | what are you specifically setting up with the Gentoo build type? |
| 02:58.22 | starseeker | we generally have Debug and Release - I haven't tried feeding in something else |
| 02:59.30 | dli | starseeker, only debug and release, if USE="debug" it's Debug, otherwise Release |
| 02:59.49 | starseeker | you're passing in CMAKE_BUILD_TYPE=Gentoo though |
| 03:00.25 | dli | starseeker, it must be from the gentoo default :( need to disable that then |
| 03:00.42 | starseeker | dli: well, it might work - just completely untested |
| 03:01.21 | starseeker | dli: this is closer, given our current state: http://pastebin.mozilla.org/1300959 |
| 03:01.31 | starseeker | you can see what happens with that |
| 03:02.07 | starseeker | I will test with a "nonsense" build type setting and see what happens once... |
| 03:03.29 | starseeker | ah, that's why the optimize and debug flags were not set |
| 03:04.15 | dli | starseeker, I can try to set default build type to Release, then, if USE=Debug, set it to Debug |
| 03:04.44 | starseeker | dli: I wouldn't worry too much about the pdf doc building just yet - if you still want to enable it, the options are BRLCAD_EXTRADOCS_PDF and BRLCAD_EXTRADOCS_PDF_MAN |
| 03:04.54 | starseeker | remember Apache FOP is needed for that |
| 03:05.24 | starseeker | dli: well, a non-standard build type should still do SOMETHING sane |
| 03:05.43 | starseeker | I'm not sure how successful we'd be fighting with the gentoo guys about that |
| 03:06.09 | dli | starseeker, BRLCAD_EXTRADOCS_PDF and BRLCAD_EXTRADOCS_PDF_MAN were already in the package script |
| 03:06.40 | starseeker | ah, k |
| 03:07.10 | dli | starseeker, the complete gentoo package (ebuild) for the svn live version: http://paste.pocoo.org/show/458500/ |
| 03:10.34 | CIA-62 | BRL-CAD: 03starseeker * r45983 10/brlcad/trunk/misc/CMake/BRLCAD_Util.cmake: Er, oops - fix default fallback settings for auto_option macro |
| 03:10.55 | starseeker | dli: that should help |
| 03:11.11 | starseeker | or at least, the settings are saner - trying the build now |
| 03:11.58 | starseeker | yeah, this should fall back to the /usr/brlcad path, no optimization and no debug flags by default |
| 03:13.49 | starseeker | note that tcl/tk need to be 8.5 - 8.4 will no longer be enough, IIRC (ttk widgets, among other issues) |
| 03:14.35 | starseeker | tnt and jama are headers, and if I'm remembering right we need our local copies |
| 03:14.46 | starseeker | shouldn't be a build issue at all |
| 03:15.09 | starseeker | (as far as conflicting with a system install anyway, unless we're copying those into our install tree - I'd need to check.) |
| 03:18.39 | dli | starseeker, building, let's see how it works out |
| 04:02.55 | starseeker | dli: any luck? |
| 18:08.28 | *** join/#brlcad ibot (~ibot@rikers.org) |
| 18:08.28 | *** topic/#brlcad is BRL-CAD Open Source Solid Modeling || http://brlcad.org || http://sf.net/projects/brlcad || #brlcad logs: http://ibot.rikers.org/%23brlcad/ || BRL-CAD release 7.20.2 is posted (20110701) || BRL-CAD is participating in the ESA Summer of Code in Space! |
| 18:10.43 | CIA-62 | BRL-CAD: 03starseeker * r45990 10/brlcad/trunk/src/archer/archer: |
| 18:10.43 | CIA-62 | BRL-CAD: Warn if we appear to be running archer from a non-install directory with an |
| 18:10.43 | CIA-62 | BRL-CAD: install already in place - this is the situation where .tcl files will get |
| 18:10.43 | CIA-62 | BRL-CAD: pulled from /usr/brlcad/* instead of local copies, which if unnoticed by the |
| 18:10.43 | CIA-62 | BRL-CAD: developer makes for some very frustrating troubleshooting (particularly for |
| 18:10.44 | CIA-62 | BRL-CAD: those not familiar with bu_brlcad_root's behavior). We have enough information |
| 18:10.45 | CIA-62 | BRL-CAD: to spot this situation at runtime, so do it. |
| 18:12.54 | bhinesley | starseeker: alright. Their verbage was a bit strong/misleading. |
| 18:18.49 | starseeker | bhinesley: they have to write that for borderline cases where a student might be trying to sprint right at the end and shoves so much code at a mentor right before the eval that they *can't* properly evaluate it |
| 18:19.43 | bhinesley | ah |
| 19:06.42 | kunigami1 | just wondering: is it feasible to pack boost compressed in the osl code? the 7ziped version has ~40MB |
| 19:07.14 | starseeker | kunigami1: um. you couldn't extract the parts you need? |
| 19:07.47 | kunigami1 | I still can't. I tried with 1.46.1. I'm currently trying with 1.47.0 |
| 19:08.24 | kunigami1 | I asked in boost dev list, but the guy suggested to me exactly what I did |
| 19:09.32 | CIA-62 | BRL-CAD: 03tbrowder2 * r45991 10/brlcad/trunk/sh/conversion.sh: add a SAVE option to keep converted tgm from being deleted |
| 19:12.38 | CIA-62 | BRL-CAD: 03tbrowder2 * r45992 10/brlcad/trunk/NEWS: documented change for users |
| 20:22.43 | *** join/#brlcad abhi2011 (~chatzilla@ip170-79-211-87.adsl2.static.versatel.nl) |
| 20:30.48 | *** join/#brlcad merzo (~merzo@137-79-132-95.pool.ukrtel.net) |
| 21:05.07 | kunigami1 | good, seems that was a bug with boost 1.46. let me see if oiio and osl compile with this smaller version |
| 21:15.56 | plaes | gotta love git-send-email \o/ |
| 21:25.58 | *** join/#brlcad KimK_ibmlaptop (~kkirwan@209.248.147.2.nw.nuvox.net) |
| 22:08.59 | CIA-62 | BRL-CAD: 03starseeker * r45993 10/brlcad/trunk/misc/CMake/CompilerFlags.cmake: Don't try adding the debug flag manually if it's Xcode - let Xcode itself handle things. Trying it manually caused an options conflict. |
| 22:55.28 | *** join/#brlcad kunigami (~kunigami@201.53.206.27) |
| 23:03.09 | CIA-62 | BRL-CAD: 03starseeker * r45994 10/brlcad/trunk/src/tclscripts/rtwizard/lib/PictureTypeBase.itcl: rt isn't always in the path. Use the bu_brlcad_root, rtwizard. |
| 23:07.40 | CIA-62 | BRL-CAD: 03starseeker * r45995 10/brlcad/trunk/NEWS: |
| 23:07.40 | CIA-62 | BRL-CAD: rtwizard was relying on rt being present in the system path in order to run it |
| 23:07.40 | CIA-62 | BRL-CAD: and generate a picture - instead, have it find and use the rt command specific |
| 23:07.40 | CIA-62 | BRL-CAD: to its own BRL-CAD release without needing rt to be in the path. |
| 23:15.24 | *** join/#brlcad merzo (~merzo@250-21-132-95.pool.ukrtel.net) |
| 23:17.12 | bhinesley | how do I get the natural origins of primitives? |
| 23:17.20 | bhinesley | cries uncle |
| 23:22.40 | CIA-62 | BRL-CAD: 03starseeker * r45996 10/brlcad/trunk/src/tclscripts/rtwizard/lib/PictureTypeBase.itcl: |
| 23:22.41 | CIA-62 | BRL-CAD: Wow. Apparently the only reason rtwizard wouldn't output png ouput if you |
| 23:22.41 | CIA-62 | BRL-CAD: specified a .png extension was due to its regarding the filename as a |
| 23:22.41 | CIA-62 | BRL-CAD: framebuffer (-F) target instead of an output file (-o). Literally a one |
| 23:22.41 | CIA-62 | BRL-CAD: character fix, and rtwizard will now output png files if given a .png filename. |
| 23:22.57 | starseeker | bhinesley: "natural origins?" |
| 23:23.14 | bhinesley | yes, that's what brlcad called it |
| 23:23.37 | bhinesley | as opposed to the bounding box center... he said that there is a natural starting point of primitives |
| 23:23.39 | starseeker | uh. is that the point that (say) the sed command grabs from a primitive? |
| 23:23.42 | starseeker | oh |
| 23:24.22 | bhinesley | I'm not sure, I'll have to look at sed |
| 23:24.35 | starseeker | would suggest checking what sed does |
| 23:24.46 | starseeker | that'd be my fist step anyway |
| 23:24.50 | starseeker | first step even |
| 23:25.08 | starseeker | (probably be shaking a fist at it too, come to think of it...) |
| 23:25.23 | bhinesley | heh |
| 23:26.10 | starseeker | look for f_sed in chgview.c in mged |
| 23:26.20 | bhinesley | ok |
| 23:27.23 | starseeker | hmm... that doesn't help much... |
| 23:28.45 | bhinesley | I guess that's the problem, starseeker; I don't really know what I'm looking for |
| 23:29.06 | starseeker | bhinesley: that's a little outside my expertice also - let me dig a minute |
| 23:29.20 | bhinesley | alright, thank you |
| 23:33.29 | starseeker | you might try looking at curr_e_axes_pos being set in edsol.c |
| 23:35.14 | starseeker | ah hah! |
| 23:35.23 | starseeker | get_solid_keypoint, edsol.c 9222 |
| 23:36.38 | bhinesley | looking |
| 23:37.09 | bhinesley | looks promising; I was thinking that it would have to be different for each primitive type |
| 23:37.24 | bhinesley | retrieving it, I mean |
| 23:38.19 | starseeker | winces - OK, if that's actually it, the options are 1) make a libged version of that 2) refactor it into per-primitive calls via the functable (a better way but more time-consuming) 3) ask brlcad what to do (the best way but probably the most work :-P) |
| 23:38.44 | bhinesley | hahaha @3 |
| 23:39.30 | starseeker | I'd say start with 1) for now (lord knows we've got lots of logic like that in there that needs cleanup) and see if you can get it to actually work |
| 23:39.57 | starseeker | if that's what we need, then it'll be already isolated from mged and easier to use for the "right thing" refactoring step |
| 23:42.21 | bhinesley | ok, it doesn't sound so bad |
| 23:50.57 | brlcad | starseeker: regarding r45996, does rtwizard's preview option still work? |
| 23:51.51 | starseeker | I believe so... |
| 23:52.33 | starseeker | tests |
| 23:52.50 | starseeker | oh, wait - point |
| 23:52.54 | starseeker | it might not... |
| 23:54.12 | starseeker | crud, yeah no dice |