| 00:02.30 | starseeker | brlcad: I can't get it to enter the bad state again, but I just managed to see that that variable was on before I lost the other debug session |
| 00:02.44 | starseeker | pretty sure that's it - it would make sense |
| 00:07.57 | brlcad | what variable? |
| 00:08.03 | starseeker | db_search_isoutput |
| 00:08.14 | brlcad | static? |
| 00:08.27 | starseeker | checks... |
| 00:08.49 | brlcad | shouldn't really be anything static in the code pushed up into librt, or even the code in libged really |
| 00:08.49 | starseeker | no |
| 00:09.01 | brlcad | as that implies statefulness |
| 00:09.12 | starseeker | search.c in librt, line 103 |
| 00:09.44 | starseeker | it does control state during the building of the plan, iirc |
| 00:10.12 | brlcad | ahh, worse! |
| 00:10.13 | brlcad | global |
| 00:10.18 | brlcad | that's statefulness |
| 00:10.21 | starseeker | nods |
| 00:10.29 | starseeker | that's been thre |
| 00:10.33 | starseeker | s/thre/there |
| 00:10.36 | starseeker | from the get-go |
| 00:11.05 | starseeker | so probably we never stressed the other search implementation enough to spot it, or we got lucky and avoided any badness accidentally |
| 00:11.24 | brlcad | you have full controll of the calling params, should be easy to eliminate |
| 00:11.39 | starseeker | you mean pass it around? |
| 00:11.48 | brlcad | better than a global |
| 00:12.00 | brlcad | especially in a lib |
| 00:12.01 | starseeker | can do that, but it means changing a lot of function arg lists to match a new template |
| 00:12.08 | starseeker | nods |
| 00:12.44 | starseeker | I can do it, but not tonight - I'm a half hour late now :-/ |
| 00:12.57 | brlcad | at a minimum, the global could be changed to be a static scope global, so it's only accessible via those functions |
| 00:13.01 | brlcad | nods |
| 00:13.11 | brlcad | no worries, we have at least a couple days of debugging |
| 00:13.23 | starseeker | gqa blew up too... augh |
| 00:13.55 | starseeker | making sure the formation of the plan starts with it at zero should do for now, and next week I'll eliminate the global |
| 00:14.03 | brlcad | suspect gqa will be easy |
| 00:14.35 | starseeker | at least the pipe fix was kinda nifty - nice work on that (you and d_rossberg) |
| 00:14.59 | starseeker | (and ``Erik for suggesting clamp) |
| 00:16.48 | starseeker | while I'm at it, I'll try to get search . and search / to work - could cheat and just supply -name * by default if no plan is available |
| 00:17.21 | starseeker | or probably just -print |
| 00:18.01 | starseeker | actually, come to think of it that should be the result of calling db_search_formplan with an empty argv list - will have to confirm that |
| 00:18.15 | starseeker | heads out |
| 00:18.39 | brlcad | yeah, I suspect it'll just happen if you specify no plan |
| 00:29.04 | CIA-14 | BRL-CAD: 03brlcad * r43714 10/brlcad/trunk/src/librt/db5_types.c: put bu_vls_true() to use so that we can more robustly detect true/false string values for standard attributes. note that the bu_str_true() will match 'R' as true too since only clearly negative values constitute false. |
| 00:30.48 | CIA-14 | BRL-CAD: 03bob1961 * r43715 10/brlcad/trunk/src/libged/nirt.c: Fixed a bug in libged/nirt.c that was tripping in the /* skip commands */ section of the windows specific code when an object being passed to nirt contained -e in its name. |
| 00:36.29 | brlcad | pretty awesome: http://www.engineeringtv.com/video/Opposed-Piston-Opposed-Cylinder |
| 00:44.06 | CIA-14 | BRL-CAD: 03brlcad * r43716 10/brlcad/trunk/src/librt/db5_types.c: ws cleanup |
| 00:50.28 | *** join/#brlcad Ralith (~ralith@S010600221561996a.vc.shawcable.net) |
| 01:57.29 | *** join/#brlcad dloman_ (~claymore@BZ.BZFLAG.BZ) |
| 01:58.33 | *** join/#brlcad tofu (~sean@BZ.BZFLAG.BZ) |
| 02:30.09 | *** join/#brlcad louipc (~louipc@archlinux/trusteduser/louipc) |
| 03:44.09 | starseeker | tofu: looks like the pipe fix is confirmed |
| 03:47.43 | CIA-14 | BRL-CAD: 03starseeker * r43717 10/brlcad/branches/cmake/ (10 files in 5 dirs): MFC r43716 |
| 04:23.47 | starseeker | aaaand libredwg doesn't get to first base |
| 04:30.28 | starseeker | growl... http://www.opendesign.com/teigha_viewer can view 'em but not export 'em |
| 04:31.29 | brlcad | starseeker: those are prime for import into autocad or proe then export as iges, step, and dxf |
| 04:32.42 | starseeker | nods - proe can read them and export them as a number of things (including pdf, which is probably most immediately useful since they aren't solid) but it'll be a long slog to do 180 one by one (especially into multiple formats) |
| 04:33.08 | starseeker | installed rpm on gentoo just to get that viewer, too... grumble |
| 04:33.18 | brlcad | iirc, proe on the linux side can be scripted |
| 04:52.14 | CIA-14 | BRL-CAD: 03brlcad * r43718 10/brlcad/trunk/src/libbu/booleanize.c: |
| 04:52.14 | CIA-14 | BRL-CAD: expand on bu_str_true() so that we can distinguish more strongly between values |
| 04:52.14 | CIA-14 | BRL-CAD: that seem to be false or true and 'everything else'. Unrecognized responses |
| 04:52.14 | CIA-14 | BRL-CAD: still return as true, but should return as >1 so callers can distinguish yes/no |
| 04:52.14 | CIA-14 | BRL-CAD: from possible input exceptions. return the first character as the exception |
| 04:52.15 | CIA-14 | BRL-CAD: value for posterity. |
| 04:52.49 | CIA-14 | BRL-CAD: 03brlcad * r43719 10/brlcad/trunk/include/bu.h: document that callers can distinguish potential exceptions with return values >1 |
| 05:24.33 | CIA-14 | BRL-CAD: 03brlcad * r43720 10/brlcad/trunk/src/mged/cmd.c: even with the plain wrapper, if the ged func returns GED_MORE, respect its authoritah |
| 06:09.56 | CIA-14 | BRL-CAD: 03brlcad * r43721 10/brlcad/trunk/src/mged/cmd.c: (log message trimmed) |
| 06:09.56 | CIA-14 | BRL-CAD: implement some nifty redraw code that looks at what objects are displayed and |
| 06:09.56 | CIA-14 | BRL-CAD: what command-line arguments were specified. if displayed objects are being |
| 06:09.56 | CIA-14 | BRL-CAD: listed on the command line, then they are potentially being edited so go ahead |
| 06:09.57 | CIA-14 | BRL-CAD: and redraw them. while this might cause redraw lag for huge models on commands |
| 06:09.57 | CIA-14 | BRL-CAD: that are merely read-only, the more common problem is the display not being |
| 06:09.58 | CIA-14 | BRL-CAD: refreshed when it should for commands that DO edit the geometry, leading to user |
| 06:14.10 | CIA-14 | BRL-CAD: 03brlcad * r43722 10/brlcad/trunk/NEWS: |
| 06:14.10 | CIA-14 | BRL-CAD: with the automatic redraw code now in the base wrapper, about 120 mged commands |
| 06:14.10 | CIA-14 | BRL-CAD: will potentially refresh the display that would have otherwise including dozens |
| 06:14.10 | CIA-14 | BRL-CAD: that modify geometry. display updates on changes should improve usability, |
| 06:14.11 | CIA-14 | BRL-CAD: reduce confusion, and help mged be more consistent |
| 06:18.25 | CIA-14 | BRL-CAD: 03brlcad * r43723 10/brlcad/trunk/src/libged/mater.c: (log message trimmed) |
| 06:18.25 | CIA-14 | BRL-CAD: restore incremental prompting functionality to the mater command. even better |
| 06:18.25 | CIA-14 | BRL-CAD: than before. leverage GED_MORE for additional prompting for any unspecified |
| 06:18.25 | CIA-14 | BRL-CAD: arguments. also takes advantage of bu_str_true()'s ability to detect input |
| 06:18.25 | CIA-14 | BRL-CAD: exceptions. this feature was prompted by discussions with ezzieyguywuf via IRC |
| 06:18.25 | CIA-14 | BRL-CAD: and having run into the problem myself several times, where mater |
| 06:18.26 | CIA-14 | BRL-CAD: unintentionally stopped prompting for missing arguments when it was migrated to |
| 06:23.41 | CIA-14 | BRL-CAD: 03brlcad * r43724 10/brlcad/trunk/NEWS: |
| 06:23.41 | CIA-14 | BRL-CAD: restored interactive prompting for mged's 'mater' command, which was |
| 06:23.41 | CIA-14 | BRL-CAD: functionality lost during the migration to libged.this feature was prompted |
| 06:23.41 | CIA-14 | BRL-CAD: after discussions with ezzieyguywuf via IRC and after having run into the |
| 06:23.41 | CIA-14 | BRL-CAD: problem myself several times while modeling. tutorial documentation was |
| 06:23.41 | CIA-14 | BRL-CAD: inconsistent as it relied on the interactive prompting for instruction. |
| 06:24.55 | *** join/#brlcad Stattrav (~Stattrav@122.167.249.60) |
| 06:24.55 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 06:52.55 | CIA-14 | BRL-CAD: 03brlcad * r43725 10/brlcad/trunk/src/libged/mater.c: restore the previous mater command behavior to report the current values being set when prompting the user for new values. approximate the old text not worry too much about getting a perfect match. |
| 06:59.31 | CIA-14 | BRL-CAD: 03brlcad * r43726 10/brlcad/trunk/src/libged/mater.c: go ahead and expand the green and blue component values. also make it clear that the values being shown are the current values. |
| 07:04.01 | *** join/#brlcad Stattrav (~Stattrav@122.167.168.51) |
| 07:04.01 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 07:59.27 | CIA-14 | BRL-CAD: 03brlcad * r43727 10/brlcad/trunk/src/libged/mater.c: |
| 07:59.28 | CIA-14 | BRL-CAD: further efforts to mime the old behavior or better. make the g and b color |
| 07:59.28 | CIA-14 | BRL-CAD: components optional once again if the color is being deleted. add back support |
| 07:59.28 | CIA-14 | BRL-CAD: for deleting the shader string and color setting. use '.' to skip parameters |
| 07:59.28 | CIA-14 | BRL-CAD: (here we deviate from original since we can't yet set default more parameters. |
| 08:00.25 | *** join/#brlcad epileg (~epileg@unaffiliated/epileg) |
| 08:02.00 | CIA-14 | BRL-CAD: 03brlcad * r43728 10/brlcad/trunk/src/libged/mater.c: fix the prompt if we're skipping/deleting the color, account for an offset of two. |
| 08:02.10 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 08:10.22 | *** part/#brlcad cjdevlin (~devlin@d118-75-252-178.try.wideopenwest.com) |
| 08:17.43 | CIA-14 | BRL-CAD: 03brlcad * r43729 10/brlcad/trunk/src/libged/mater.c: offset pushing towards the wrong direction, be positive. also catch the error case where color has been deleted but then later is 'skipped'. |
| 08:25.46 | *** join/#brlcad cjdevlin (~devlin@d118-75-252-178.try.wideopenwest.com) |
| 08:29.56 | CIA-14 | BRL-CAD: 03brlcad * r43730 10/brlcad/trunk/src/libged/mater.c: tweak the usage, let user know which channel is teh suck |
| 08:30.36 | CIA-14 | BRL-CAD: 03brlcad * r43731 10/brlcad/trunk/doc/docbook/lessons/en/mged11_refining_mug.xml: update tutorial example interactive output for the mater command to reflect new form |
| 08:32.15 | CIA-14 | BRL-CAD: 03brlcad * r43732 10/brlcad/trunk/TODO: mater now properly prompts for missing parameters interactively |
| 08:35.51 | *** join/#brlcad Stattrav (~Stattrav@122.172.15.235) |
| 08:35.51 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 08:37.26 | CIA-14 | BRL-CAD: 03brlcad * r43733 10/brlcad/trunk/ (NEWS TODO): |
| 08:37.26 | CIA-14 | BRL-CAD: bob reportedly fixed the bug causing nirt to not work on windows. user reported |
| 08:37.26 | CIA-14 | BRL-CAD: that nirt on windows was failing if run in a command window. bob found a |
| 08:37.26 | CIA-14 | BRL-CAD: problem where it was getting tripped up on objects with '-e' in their name. |
| 09:22.29 | *** join/#brlcad alex_joni (~alex_joni@emc/board-of-directors/alexjoni) |
| 09:57.52 | *** join/#brlcad Ralith_ (~ralith@S010600221561996a.vc.shawcable.net) |
| 11:13.16 | CIA-14 | BRL-CAD: 03jordisayol * r43734 10/brlcad/trunk/sh/ (make_deb.sh make_rpm.sh): update copyright to 2011 |
| 13:11.14 | starseeker | makes a note to check into Pro/Batch |
| 13:30.42 | starseeker | yeah... the Tegiha file convertor worked under wine, and librecad can open the dxf files (so can we with dxf-g) but there aren't any good ways I can find to get decent pdf output - gonna have to use Pro/Batch or something like it |
| 13:46.46 | starseeker | O.o freecad is in main portage now? (wrong category, media-gfx, but still...) |
| 13:47.16 | starseeker | tries it... haven't had great luck with the sci overlay builds to date... |
| 14:34.06 | CIA-14 | BRL-CAD: 03starseeker * r43735 10/brlcad/branches/cmake/src/librt/ (search.c search.h): |
| 14:34.06 | CIA-14 | BRL-CAD: Globals are evil - need to pass db_search_isoutput around as a parameter during |
| 14:34.06 | CIA-14 | BRL-CAD: the process of forming the plan. As a side benefit, it shouldn't be necessary |
| 14:34.06 | CIA-14 | BRL-CAD: for f_print to reset the value anymore, since the scope of the variable is now |
| 14:34.06 | CIA-14 | BRL-CAD: limited to the plan forming process |
| 14:34.30 | starseeker | brlcad: I think that's got it |
| 14:34.44 | starseeker | still need to fix behavior for search . and friends |
| 14:50.27 | *** join/#brlcad piksi (piksi@pi-xi.net) |
| 15:01.40 | *** join/#brlcad piksi (piksi@pi-xi.net) |
| 15:03.31 | *** join/#brlcad piksi (piksi@pi-xi.net) |
| 15:03.46 | *** part/#brlcad piksi (piksi@pi-xi.net) |
| 15:40.28 | ezzieyguywuf | wow, brlcad you've been busy. |
| 15:40.36 | ezzieyguywuf | I still need to submit that bug report... |
| 16:04.03 | ezzieyguywuf | ah, another discrepency I noticed in the tutorials is when it tells you to cp certain shapes, it never tells you that you have to draw them first before you can do anything to them. |
| 16:31.30 | starseeker | those hessi dwg files may be even more extensive than simplesat... |
| 16:31.35 | starseeker | sweet |
| 16:48.18 | *** join/#brlcad piksi (piksi@pi-xi.net) |
| 17:35.17 | ezzieyguywuf | heh, I'm running mged vers 7.16.8 which is > vers 6.0 so according to the tute the order of inside should be botto-top-rear-front-right-left but instead I got front-rear-right-left-bottom-top |
| 17:36.50 | vtts | are all menu actions executable from command line? e.g. is it possible to enter multi pane mode, set active pane, show model axes on it and so on? |
| 17:41.44 | ezzieyguywuf | vtts: pretty sure you can. checkout the quick reference card. |
| 17:42.11 | vtts | didn't find it there |
| 17:42.14 | ezzieyguywuf | hrm. |
| 17:42.42 | vtts | i can change view options of an active pane, but don't know how tu change mode or switch to other pane |
| 17:43.30 | vtts | found a few functions in the source, but don't know where to get id from (e.g. setmv id) |
| 17:44.43 | vtts | eh... silly me... id is on the title bar |
| 17:44.59 | vtts | does a face-palm action |
| 17:45.24 | ezzieyguywuf | :-D |
| 18:04.20 | ezzieyguywuf | hrm, still no dotted lines for me.... |
| 18:33.23 | ezzieyguywuf | so man, its taking a while for the raytracer to raytrace this toy truck. sometimes it even fails. Could it be I'm doing something wrong? |
| 18:33.38 | ezzieyguywuf | I would expect perfermance for something of this simple geometry to be pretty good. |
| 18:36.08 | ezzieyguywuf | http://ompldr.org/vN29hYw screenshot of raytrace after ~1min |
| 19:40.56 | *** join/#brlcad Stattrav (~Stattrav@117.192.133.151) |
| 19:40.56 | *** join/#brlcad Stattrav (~Stattrav@unaffiliated/stattrav) |
| 20:15.05 | ezzieyguywuf | so....what are my chances of modelling an airfoil using brl-cad? |
| 22:22.24 | CIA-14 | BRL-CAD: 03jordisayol * r43736 10/brlcad/trunk/ (10 files in 3 dirs): add new GNU/Linux mime-type icons for geometry db file v.4 and v.5 |
| 23:13.39 | Ralith | Would there be any interest in an effort to add GPU accelereation to the raytracer? |
| 23:58.35 | *** join/#brlcad piksi (piksi@pi-xi.net) |