Age | Commit message (Collapse) | Author | |
---|---|---|---|
2012-12-12 | move edge split into its own function which can be called by the modifier ↵ | Campbell Barton | |
without having to call a bmesh operator (gives some speedup). | |||
2012-12-12 | use header only tags for edge split (to prepare to move this into tools/) | Campbell Barton | |
2012-12-12 | add threshold for bmesh & openmp so its not used with low poly meshes, ↵ | Campbell Barton | |
BM_OMP_LIMIT may need tweaking. | |||
2012-12-12 | use openmp to thread some common bmesh operations | Campbell Barton | |
- BM_mesh_elem_toolflags_ensure / bmo_flag_layer_alloc / bmo_flag_layer_free / bmo_flag_layer_clear - BM_mesh_select_flush - EDBM_index_arrays_init notes: - mostly use openmp `sections` to split operations on vert/edge/face since this is a fairly minor change. - split tool flag pool in 3, this means we can allocate exact sizes needed and iterate on them in threads without alloc'ing. | |||
2012-12-12 | replace BLI_array_declare with BLI_array_staticdeclare() and ↵ | Campbell Barton | |
BLI_array_alloca() for smaller arrays. | |||
2012-12-11 | replace BLI_array_fixedstack_declare with() new macro BLI_array_alloca() ↵ | Campbell Barton | |
which uses stack memory always and doesn't need to be freed explicitly. | |||
2012-12-11 | minor speedup - replace use of smallhash with api_flags for BM_edge_split() | Campbell Barton | |
2012-12-11 | own cleanup commit in bmesh branch - removed last letters from ends of some ↵ | Campbell Barton | |
comments. | |||
2012-12-08 | bmesh toolflags would use BLI_MEMPOOL_SYSMALLOC when reducing layers only ↵ | Campbell Barton | |
(would use MEM_mallocN in one case but malloc in another). better use blenders MEM_mallocN for both. | |||
2012-12-08 | fix [#33438] Bevel modifier "angle" mode is broken | Campbell Barton | |
bevel modifier was making zero area faces & edges that made scanfill fail (since it no longer removes doubles when filling ngons) | |||
2012-12-07 | Bevel: fix 'causing artifacts' bug 33245. | Howard Trickey | |
Really was caused by a previous bevel making a two-edged face, which caused other faces to be dropped when copying a bmesh. The quadstrip code needed to be more careful to avoid creating two-edge faces. | |||
2012-12-04 | bevel - use tri-fan filling in the special case when a bevel edge meets a ↵ | Campbell Barton | |
non bevel edge at a valence 2 vert. this is the topology tri-fan was intended to be used. | |||
2012-12-04 | bevel: re-order checks so angle checks are done after quick sanity checks. | Campbell Barton | |
2012-12-03 | revert part of r52720, Id rather leave these as-is, even if they give ↵ | Campbell Barton | |
warnings under some configurations. | |||
2012-12-03 | fix [#33392] In-dev freeway generation addon crashes on recent builds. | Campbell Barton | |
2012-12-03 | use const for bm_mesh_allocsize_default, bm_mesh_chunksize_default | Campbell Barton | |
2012-12-03 | bridge tool - simple optimization, break early if edge loop length ↵ | Campbell Barton | |
comparisons are worse then existing best loop test. | |||
2012-12-03 | fix [#33391] Bridge two Edgeloops fails in simple case | Campbell Barton | |
2012-12-03 | Bevel: fix for bulging part of bug 33280. | Howard Trickey | |
Bulging still happens, but fixed the cases where it was obvious because it destroys an otherwise straight 'pipe' by snapping the vertex mesh points to that pipe. | |||
2012-12-02 | Silent a bunch of gcc warnings (usually dummy, but noisy!). | Bastien Montagne | |
2012-12-02 | Silent some warnings (the one in bmesh_operator.c was even preventing build ↵ | Bastien Montagne | |
in -Werror mode). | |||
2012-12-01 | Bevel: fix crash bug 33362, when beveling one edge at valence 2 vertex. | Howard Trickey | |
That special case should have been tested before - the code was wrong in about three different ways. | |||
2012-12-01 | Bevel: fix spike in suzanne, bug 33354. | Howard Trickey | |
Non-planar faces made some of the meet point code not work well, so now calculate local face norms. | |||
2012-11-30 | fix for various asserts running operator tests - not likely any of these ↵ | Campbell Barton | |
would cause real user bugs though. | |||
2012-11-30 | fix for another error in my own recent commit, beauty fill crashed. | Campbell Barton | |
2012-11-30 | fix own mistake in recent commit - entering editmode with a shape key crashed. | Campbell Barton | |
2012-11-29 | remove unneeded mesh->bmesh conversion code that copied BMLoop data in a ↵ | Campbell Barton | |
separate loop (since there is no longer pre-allocated loop data on the new BMLoops). | |||
2012-11-29 | fix [#33029] Applying modifier leaks memory | Campbell Barton | |
Thanks for Sergey for finding the bug & patching, This fix works a bit differently. Theres no need to allocate the customdata in the first place - since its written into. So add a flag for vert/edge/face/loop creation functions so they can skip customdata creation. | |||
2012-11-29 | Bevel: partial fix for distortion (bug 33280). | Howard Trickey | |
Sometimes it is impossible to have same offset from both edges that are on angled faces. The fix here at least doesn't distort the non-beveled part of the model, and looks much better than before on bug example, but is still not perfect. | |||
2012-11-29 | bmesh py api: add bmesh.update_edit_mode(), there was no way to redraw the ↵ | Campbell Barton | |
3d view or re-calculate face tessellation from python. add py template for editing meshes in editmode. also remove double call to CTX_wm_region which does a string lookup. | |||
2012-11-28 | fix for minor annoyance - when extruding a circle there would always be one ↵ | Campbell Barton | |
face flipped, now edges are all ordered in the same direction. | |||
2012-11-28 | typo's and some style cleanup, also added asserts into BLI_vsnprintf and ↵ | Campbell Barton | |
BLI_sprintfN when invalid args are given. | |||
2012-11-28 | Copy face attributes when creating new faces in BMO_symmetrize | Nicholas Bishop | |
Fixes bug [#33269] Symmetrize doesnt honor shadesmooth projects.blender.org/tracker/index.php?func=detail&aid=33269&group_id=9&atid=498 | |||
2012-11-28 | fix [#33320] Decimate modifer in collapse is inconsistent when limiting to a ↵ | Campbell Barton | |
vertex group | |||
2012-11-28 | use radians for 'spin' bmesh operator (since the rest of the py api uses ↵ | Campbell Barton | |
radians). also rename BMO_OP_SLOT_SUBTYPE_MAP_FLOAT -> BMO_OP_SLOT_SUBTYPE_MAP_FLT for consistency. | |||
2012-11-28 | bmesh operator naming - use clearer names for args eg: (mat -> matrix, ↵ | Campbell Barton | |
use_singleedge -> use_single_edge) also remove duplicate docs for operator arg formatting. | |||
2012-11-28 | improve docstring for BMO_op_vinitf(). | Campbell Barton | |
2012-11-27 | use clearer names for 'single' bmesh operator args & add '%e' to ↵ | Campbell Barton | |
BMO_op_vinitf comments. | |||
2012-11-27 | py/bmesh api - add support for single item buffers (odd feature but used ↵ | Campbell Barton | |
quite a bit with bmesh operators). also add utility functions BMO_slot_buffer_from_single(), BMO_slot_buffer_get_single() | |||
2012-11-27 | code cleanup: bmesh operator comments, readying for doc-generation. | Campbell Barton | |
2012-11-27 | fix for asserts added in own recent commit with more strict type-checking | Campbell Barton | |
- BMO_slot_copy now only copies compatible elements. other minor changes - don't use text.format(...), convention for UI scripts is C style string formatting. - rename bmo_edgenet_prepare --> bmo_edgenet_prepare_exec - float/double warning in bevel. | |||
2012-11-27 | Bevel: better round profile code. | Howard Trickey | |
Easier to understand, no touchy intersect code, and works even when arms aren't equal length. Old code ifdef'd for now, will remove soon. | |||
2012-11-27 | bmesh/py operator api: | Campbell Barton | |
add type checking for element buffers, there was nothing stopping python from passing any element type into an argument when in some cases only verts/edges/faces were expected. now operator args define which types they support. | |||
2012-11-27 | fix/workaround [#33281] script goes into not responding | Campbell Barton | |
scanfill remove-doubles pass assumes ordered edges (as with curves), otherwise it can hang. workaround this problem by skipping removing-doubles for mesh ngons, since this isnt such a common case as it is with curves and we can just not support it. | |||
2012-11-26 | style cleanup | Campbell Barton | |
2012-11-26 | py/bmesh api - support for converting from/to BMO_OP_SLOT_MAPPING type. | Campbell Barton | |
2012-11-26 | Fix #33285: loop cut is not supposed to cut through triangles/ngons, but it | Brecht Van Lommel | |
still happened when the loop would go all the way around the mesh with just one triangle/ngon inbetween to close the loop. | |||
2012-11-26 | use slightly more efficient BLI_array_fixedstack_declare() instead of ↵ | Campbell Barton | |
BLI_array_staticdeclare() for bevel code since the array size is known. | |||
2012-11-26 | fix for uninitialized memory use with numeric input: | Campbell Barton | |
bevel/inset/marker-move would use uninitialized memory when used as modal operators and pressing backspace after entering values. | |||
2012-11-26 | use more rigid type checking for bmesh slot subtypes. | Campbell Barton | |