| 01:08.47 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 01:52.33 | *** join/#brlcad teepee_ (~teepee@unaffiliated/teepee) |
| 03:33.36 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 07:06.13 | *** join/#brlcad Caterpillar (~caterpill@unaffiliated/caterpillar) |
| 09:58.48 | *** join/#brlcad merzo_ (~merzo@94-75-132-95.pool.ukrtel.net) |
| 12:48.58 | *** join/#brlcad teepee_ (~teepee@unaffiliated/teepee) |
| 13:27.47 | Notify | 03BRL-CAD:starseeker * 69074 (brlcad/trunk/include/rt/nurb.h brlcad/trunk/src/librt/CMakeLists.txt and 23 others): More logic moving and header trimming in preparation for library separation. |
| 14:21.28 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 14:28.53 | Notify | 03BRL-CAD:starseeker * 69075 (brlcad/trunk/include/nmg.h brlcad/trunk/include/rt/nmg.h and 6 others): I doubt this whole re_nmgfree list will survive long term, but in the short term it's coupling nmg to the RTG global. Make it an nmg.h global. Also start roughing out the nessary bits to get the ray intersection test needed by nmg_class to be self contained. |
| 15:01.03 | Notify | 03BRL-CAD:starseeker * 69076 (brlcad/trunk/include/nmg.h brlcad/trunk/include/rt/nmg.h brlcad/trunk/src/librt/primitives/nmg/nmg.c): We're going to need an adapter layer if we want to separate this - it looks like nmg_isect_ray_model is probably the best point. |
| 16:07.51 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 16:17.19 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 16:32.54 | *** join/#brlcad KimK (~Kim__@2600:8803:7a85:6d00:a907:e127:a878:d60a) |
| 17:31.39 | *** join/#brlcad merzo_ (~merzo@252-121-133-95.pool.ukrtel.net) |
| 18:36.26 | *** join/#brlcad kintel (~kintel@unaffiliated/kintel) |
| 20:06.30 | Notify | 03BRL-CAD:starseeker * 69077 (brlcad/trunk/include/nmg.h brlcad/trunk/include/rt/nmg.h and 5 others): Punting slightly by making the nmg and rt versions of structs castable, but this technically breaks the dependency of nmg_rt_isect on rt/nmg.h. More extensive rework to make nmg properly independent of the form of the librt structs will probably involve narrowing the scope of nmg_rt_isect to returning an ordered |
| 20:06.32 | Notify | list of hit points - need to study what the actual requirements of nmg_class are. That logic untangling and rework looks like it will be a singificant effort in its own right; in the meantime this should serve the immediate purpose. |
| 20:06.34 | Notify | ... |
| 20:28.21 | Notify | 03BRL-CAD:starseeker * 69078 brlcad/trunk/src/librt/primitives/nmg/nmg_class.c: completely remove rt/nmg.h from nmg_class.c |
| 20:44.14 | Notify | 03BRL-CAD:starseeker * 69079 brlcad/trunk/src/libbrep/boolean.cpp: looks like we don't actually need raytrace.h in here (and we shouldn't, since librt is supposed to use libbrep not the other way around...) |
| 21:01.13 | Notify | 03BRL-CAD:starseeker * 69080 (brlcad/trunk/include/nmg.h brlcad/trunk/include/rt/CMakeLists.txt and 66 others): merge rt/nurb.h into nmg.h |
| 21:04.18 | Notify | 03BRL-CAD:n_reed * 69081 (brlcad/branches/brep-debug/INSTALL brlcad/branches/brep-debug/NEWS and 209 others): sync r68823 through 68945 changes from trunk |
| 21:07.04 | Notify | 03BRL-CAD:starseeker * 69082 brlcad/trunk/src/librt/CMakeLists.txt: put oslo files with the rest of the nmg nurbs logic in bspline. |
| 21:11.46 | Notify | 03BRL-CAD:starseeker * 69083 (brlcad/trunk/include/nmg.h brlcad/trunk/src/librt/bezier.c and 3 others): bezier functions group logically with the rest of bspline... |
| 21:13.11 | Notify | 03BRL-CAD:starseeker * 69084 brlcad/trunk/src/conv/iges/trimsurf.c: shadowed global. Note - need to check and see if this logic relates to or belongs in a lower lib... |
| 21:19.43 | Notify | 03BRL-CAD:starseeker * 69085 (brlcad/trunk/src/librt/CMakeLists.txt brlcad/trunk/src/librt/primitives/bspline/nurb_bezier.c): merge bezier.c into nurb_bezier.c |
| 21:21.13 | Notify | 03BRL-CAD:starseeker * 69086 (brlcad/trunk/src/librt/primitives/bspline/nurb_oslo_calc.c brlcad/trunk/src/librt/primitives/bspline/nurb_oslo_map.c): remove raytrace.h from the oslo files |
| 21:25.25 | Notify | 03BRL-CAD:starseeker * 69087 brlcad/trunk/src/librt/CMakeLists.txt: nurb_brep.cpp is not a candidate file for mergeing into libnmg - rename for clarity. |
| 21:26.15 | Notify | 03BRL-CAD:starseeker * 69088 brlcad/trunk/src/librt/primitives/bspline/bspline_brep.cpp: fix header |
| 21:27.44 | Notify | 03BRL-CAD:starseeker * 69089 brlcad/trunk/src/librt/db_diff.c: db_diff.c doesn't appear to need the nmg.h include |
| 21:30.56 | *** join/#brlcad merzo_ (~merzo@93-195-113-92.pool.ukrtel.net) |
| 21:45.39 | starseeker | achieves a successfull separation of libnmg in a test build |
| 21:45.46 | starseeker | whew - finally |
| 21:47.00 | starseeker | had decided to wait til after release, but now that it looks like it might actually work I'm tempted... |
| 21:51.45 | Notify | 03BRL-CAD:starseeker * 69090 (brlcad/trunk/include/bu/exit.h brlcad/trunk/src/libanalyze/MeshHealing/DCEL.h): fix regress-repository |
| 22:14.34 | Notify | 03BRL-CAD:starseeker * 69091 (brlcad/trunk/include/nmg.h brlcad/trunk/src/CMakeLists.txt brlcad/trunk/src/librt/CMakeLists.txt): Create an independent libnmg, moving the relevant logic out of librt. This is the culmination of the series of refactorings that have been taking place in librt. This should be reverted if it looks like it will cause any problems at all with 7.26.2 - I'm mostly committing it at this time |
| 22:14.36 | Notify | to more easily test on multiple platforms. |
| 22:14.38 | Notify | ... |
| 22:49.15 | Notify | 03BRL-CAD:starseeker * 69092 (brlcad/trunk/include/nmg.h brlcad/trunk/src/CMakeLists.txt brlcad/trunk/src/librt/CMakeLists.txt): Yep, not ready -r everse r69091. Windows is revealing some problems that weren't showing up on Linux. |
| 22:58.17 | Notify | 03BRL-CAD:starseeker * 69093 (brlcad/trunk/include/nmg.h brlcad/trunk/include/rt/nmg.h and 5 others): nmg_stash_model_to_file appears to be used for debugging, but its not viable in a libnmg breakout because it relies heavily on the librt db io routines. if we need this in libnmg, we'll have to come up with a different solution. |
| 23:00.32 | Notify | 03BRL-CAD:starseeker * 69094 (brlcad/trunk/src/librt/globals.c brlcad/trunk/src/librt/primitives/nmg/nmg_misc.c): whoops - forgot to move these from globals.c |
| 23:02.17 | Notify | 03BRL-CAD:starseeker * 69095 (brlcad/trunk/src/librt/globals.c brlcad/trunk/src/librt/primitives/nmg/nmg_plot.c): another globals.c stray |
| 23:05.58 | Notify | 03BRL-CAD:starseeker * 69096 (brlcad/trunk/include/nmg.h brlcad/trunk/src/librt/primitives/sketch/sketch.c): sketch uses this - properly declare it. |
| 23:49.46 | starseeker | OK, Windows built with those changes. |
| 23:50.03 | starseeker | (or rather, those changes hacked into the breakout libnmg build) |
| 23:55.59 | starseeker | wonders if the libg breakout will be more or less painful... |