| 01:34.02 | *** join/#brlcad LordOfBikes_ (~armin@dslb-188-098-168-050.188.098.pools.vodafone-ip.de) |
| 03:17.34 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 08:27.27 | *** join/#brlcad Boquete (~piotr@aaaf110.neoplus.adsl.tpnet.pl) |
| 12:25.22 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 13:10.00 | Notify | 03BRL-CAD:starseeker * 66274 (brlcad/trunk/include/brep.h brlcad/trunk/src/libbrep/shape_recognition.cpp and 4 others): After a few mistakes, I now see why openNURBS named some of their variables as they did. Make id, type, and a few other variables inside structs clearer. |
| 13:37.10 | Notify | 03BRL-CAD:starseeker * 66275 (brlcad/trunk/include/brep.h brlcad/trunk/src/libbrep/shape_recognition.cpp and 3 others): Add boolean ops for B-Reps |
| 13:37.16 | ``Erik | plays with org-trello some |
| 14:01.24 | *** join/#brlcad teepee (~teepee@unaffiliated/teepee) |
| 14:02.01 | Notify | 03BRL-CAD:starseeker * 66276 (brlcad/trunk/src/libbrep/shape_recognition.h brlcad/trunk/src/libbrep/shape_recognition_cylinder.cpp and 2 others): Add attributes listing the loops (and for islands, faces) that went into the definition of that object. Probably want to turn these off by default (maybe enable with a debug flag...) to keep the size down. |
| 14:19.03 | *** join/#brlcad alexandrus (~alexandru@p5B2EEBD0.dip0.t-ipconnect.de) |
| 14:36.38 | starseeker | gawks in disbelief at the behavior of the intersection operator |
| 14:37.01 | starseeker | if I make a comb that intersects a cylinder and an arbn, I get the expected shape |
| 14:37.37 | starseeker | if I turn around and subtract that comb from another shape, it doesn't subtract the net shape of the comb - instead it appears to be subtracting the arbn in the comb from the parent |
| 14:59.16 | Notify | 03BRL-CAD:starseeker * 66277 brlcad/trunk/src/libbrep/tests/nurbs_csg.g: add another example |
| 15:00.12 | brlcad | define a region |
| 15:01.20 | brlcad | with no regions defined, each primitive is assumed to be a region, and you have an intersection occurring above the region level causing unexpected behavior |
| 15:02.02 | brlcad | default should be to make the top-level a region, long-standing source of unexpected behavior |
| 15:03.37 | starseeker | hmm... not what I thought was happening, simple tests do the expected thing |
| 15:03.38 | brlcad | happens that way so it's at least consistent when you have models with some regions, when it traverses a path and gets to the leaf without a region it makes the leaf a region |
| 15:03.50 | starseeker | my mistake must be elsewhere |
| 15:04.33 | starseeker | probably making an arbn when I shouldn't be |
| 15:04.35 | brlcad | could be several other things, but that's what first jumps to mind -- old issue |
| 15:04.47 | starseeker | double checks... |
| 15:05.48 | Notify | 03BRL-CAD:starseeker * 66278 brlcad/trunk/src/libbrep/tests/nurbs_csg.g: add another example. |
| 15:06.21 | starseeker | yeah, looks like that's it |
| 15:06.29 | starseeker | sorry for the noise |
| 15:06.41 | starseeker | rolls up his sleves and dives back into the planar assembly code... |
| 15:07.39 | starseeker | (arbn lines from subtraction comb happened to line up with the parent shape lines - got me confused) |
| 15:11.24 | brlcad | not noise, it's something we should change :) |
| 15:11.31 | brlcad | hits almost every use eventually |
| 15:11.36 | brlcad | s/use/user/ |
| 15:12.26 | brlcad | I think I hit it my very first new model attempt |
| 15:20.33 | alexandrus | hmm...my the latest brlcad/mged crashes with lots of infinite loop and ::itcl:: pattern issues... |
| 15:20.47 | alexandrus | this is not my day |
| 15:24.39 | alexandrus | seems, only the i386 version has this... |
| 16:24.11 | *** join/#brlcad Boquete (~piotr@aaaf110.neoplus.adsl.tpnet.pl) |
| 18:42.53 | Notify | 03BRL-CAD:starseeker * 66279 (brlcad/trunk/src/libbrep/shape_recognition.cpp brlcad/trunk/src/libbrep/shape_recognition.h and 4 others): Take another stab at hierarchy handling. |
| 19:24.15 | Notify | 03BRL-CAD:starseeker * 66280 brlcad/trunk/src/libbrep/shape_recognition_cylinder.cpp: Wrong check was causing garbage rcc/arbn primitives with a number of faces in NIST2 - loop 5 is one test case. |
| 19:29.41 | *** join/#brlcad LordOfBikes (~armin@dslb-188-098-168-050.188.098.pools.vodafone-ip.de) |
| 19:59.31 | Notify | 03BRL-CAD:starseeker * 66281 brlcad/trunk/src/libbrep/shape_recognition.cpp: Don't infinite loop if we have planar island faces with inner loops. |
| 20:20.02 | Notify | 03BRL-CAD:starseeker * 66282 brlcad/trunk/src/libbrep/shape_recognition_cylinder.cpp: fix multi_cylinder_01 |
| 20:21.03 | Notify | 03BRL-CAD:starseeker * 66283 brlcad/trunk/src/libged/shape_recognition.cpp: make slightly more convenient toplevel names for tab completion. |
| 20:56.55 | Notify | 03BRL-CAD:starseeker * 66284 brlcad/trunk/src/libbu/path.c: If period_pos is null, don't try to trunc |
| 20:57.29 | Notify | 03BRL-CAD:starseeker * 66285 brlcad/trunk/src/libged/shape_recognition.cpp: get the brep name back from the brep function. |
| 21:30.42 | Notify | 03BRL-CAD:starseeker * 66286 brlcad/trunk/src/libbrep/shape_recognition.cpp: If we're going with a brep, clean up the child objects so we don't get stray CSG bits added |
| 21:37.03 | Notify | 03BRL-CAD:starseeker * 66287 (brlcad/trunk/src/libbrep/shape_recognition.h brlcad/trunk/src/libbrep/shape_recognition_util.cpp): Clean up some stray code |
| 21:44.23 | Notify | 03BRL-CAD:starseeker * 66288 brlcad/trunk/src/libbrep/shape_recognition.cpp: update comment |
| 21:44.46 | starseeker | phew. there we go, can handle all 5 nist test cases again (knock on wood) |