January 2012 - cgma-dev - mailman3.cels.anl.gov

Re: [cgma-dev] OpenCascade 6.5.2
by Jed Brown 09 Jan '12

09 Jan '12

On Mon, Jan 9, 2012 at 09:46, Jane Hu <janejhu(a)gmail.com> wrote: > Hi, Jed: > > Current cgm code is compatible to OpenCascade6.5.1 and lower. In > OpenCascade 6.5.2, the API calls all added a geometry tolerance to the > arguments, which is reasonable. We need to update our current source code > to accommodate the changes. > > I will let you know when this is done. > Great, thank you.

1 0

Re: [cgma-dev] OpenCascade 6.5.2
by Jane Hu 09 Jan '12

09 Jan '12

Hi, Jed: Current cgm code is compatible to OpenCascade6.5.1 and lower. In OpenCascade 6.5.2, the API calls all added a geometry tolerance to the arguments, which is reasonable. We need to update our current source code to accommodate the changes. I will let you know when this is done. Thank you! Jane On Sat, Jan 7, 2012 at 2:55 PM, Jed Brown <jedbrown(a)mcs.anl.gov> wrote: > I get type conversion errors when building current (r5293) with > OpenCascade 6.5.2, compiler errors attached.

1 0

OpenCascade 6.5.2
by Jed Brown 07 Jan '12

07 Jan '12

I get type conversion errors when building current (r5293) with OpenCascade 6.5.2, compiler errors attached.

1 0

Re: [cgma-dev] imprint mesh bug
by Jane Hu 06 Jan '12

06 Jan '12

Hi, Boyd: Thank you for debugging this issue for me, otherwise, it's a bit difficult to reproduce your work just in cgm. I added a function in OCCBody that returns its shape no matter of its shapetype, this way, the function will always return something. Hope it fixed your problem. Please double check on it. Thanks. Jane On Wed, Dec 28, 2011 at 2:44 PM, Boyd Tidwell <boyd(a)csimsoft.com> wrote: > > The following commands: > > brick width 10 > brick width 6 > body 2 move x 8 > volume 1 2 size 1 > mesh volume 1 2 > imprint mesh body 1 2 > > cause a crash. The same commands without the meshing works fine. > > > The actual failure occurs in > > void PartitionSurface::get_parents_virt( DLIList<TopologyBridge*>& > result_list ) > > On the assert in the following code, tmp_list.size() is zero. > > for( i = real_surf_shells.size(); i--; ) > { > tmp_list.clean_out(); > real_surf_shells.get_and_step()->get_parents_virt( tmp_list ); > assert(tmp_list.size() == 1); > real_surf_vols.append(tmp_list.get()); > } > > > A clue to the real problem may be revealed in the method called right > before the assert: > > > void OCCShell::get_parents_virt( DLIList<TopologyBridge*>& parents ) > > for(int i = 0; i < bodies->size(); i++) > { > body = bodies->get_and_step(); > TopoDS_Compound* shape = body->get_TopoDS_Shape(); <---- > added line > if (shape == > NULL) > <---- added line > > continue; > <---- added line > TopExp::MapShapesAndAncestors(*(body->get_TopoDS_Shape()), > TopAbs_SHELL, TopAbs_SOLID, M); > if (!M.Contains(*(get_TopoDS_Shell()))) > continue; > > > Previously I added the lines to check if "body->get_TopoDS_Shape()" was > returning a NULL pointer. It seemingly fixed a number of problems. I wish > I'd written down the actual things that the change fixed. This might just > be a band-aid that hides a deeper problem but my tests run much better with > the new code in place. Because the existing code doesn't check for a null > pointer, there seems to be an assumption that it will never be null so the > null pointer could be the real problem. I don't have enough understanding > of the code to know what is correct. > > Good Luck and thanks in advance, > > - Boyd >

1 0

r5293 - cgm/trunk/geom/OCC
by jiangtao_ma＠yahoo.com 06 Jan '12

06 Jan '12

Author: janehu Date: 2012-01-06 12:27:55 -0600 (Fri, 06 Jan 2012) New Revision: 5293 Modified: cgm/trunk/geom/OCC/OCCBody.cpp cgm/trunk/geom/OCC/OCCBody.hpp cgm/trunk/geom/OCC/OCCModifyEngine.cpp cgm/trunk/geom/OCC/OCCQueryEngine.cpp cgm/trunk/geom/OCC/OCCShell.cpp cgm/trunk/geom/OCC/OCCSurface.cpp Log: Fixed the bug that sometimes shell can't find its parent by adding a overload function to return body's shape no matter what type it has. Before this function only returns body's Compound shape. Passed make check in cgm. Modified: cgm/trunk/geom/OCC/OCCBody.cpp =================================================================== --- cgm/trunk/geom/OCC/OCCBody.cpp 2012-01-06 18:24:46 UTC (rev 5292) +++ cgm/trunk/geom/OCC/OCCBody.cpp 2012-01-06 18:27:55 UTC (rev 5293) @@ -81,10 +81,42 @@ TopoDS_Compound* OCCBody::get_TopoDS_Shape() { if (myTopoDSShape && !myTopoDSShape->IsNull()) + { assert(myTopoDSShape->ShapeType() == TopAbs_COMPOUND); - return myTopoDSShape; + return myTopoDSShape; + } } +void OCCBody::get_TopoDS_Shape(TopoDS_Shape *& shape) +{ + if (myTopoDSShape && !myTopoDSShape->IsNull()) + { + assert(myTopoDSShape->ShapeType() == TopAbs_COMPOUND); + shape = myTopoDSShape; + } + else + { + DLIList<Lump*> lumps = this->lumps(); + DLIList<OCCShell*> shells = this->shells(); + DLIList<OCCSurface*> surfaces = this->my_sheet_surfaces(); + if(lumps.size() + shells.size() + surfaces.size() == 1) + { + if(lumps.size() == 1) + { + OCCLump* lump = CAST_TO(lumps.get(), OCCLump); + assert(lump != NULL); + shape = CAST_TO(lump, OCCLump)->get_TopoDS_Solid(); + } + else if(shells.size() == 1) + shape = shells.get()->get_TopoDS_Shell(); + else + shape = surfaces.get()->get_TopoDS_Face(); + } + else + shape = NULL; + } +} + void OCCBody::set_sheet_surfaces(DLIList<OCCSurface*> surfaces) { mySheetSurfaces = surfaces; Modified: cgm/trunk/geom/OCC/OCCBody.hpp ===================================================================

1 0

r5292 - cgm/trunk/test
by jiangtao_ma＠yahoo.com 06 Jan '12

06 Jan '12

Author: janehu Date: 2012-01-06 12:24:46 -0600 (Fri, 06 Jan 2012) New Revision: 5292 Modified: cgm/trunk/test/imprint_bug.cpp Log: Added test for imprint mesh bug filed by Boyd Tidwell, the bug is acctually in finding shell's parent, so simply test in the file to just test the get_parents functoin. Modified: cgm/trunk/test/imprint_bug.cpp =================================================================== --- cgm/trunk/test/imprint_bug.cpp 2012-01-06 18:08:22 UTC (rev 5291) +++ cgm/trunk/test/imprint_bug.cpp 2012-01-06 18:24:46 UTC (rev 5292) @@ -16,6 +16,7 @@ #include "CubitMessage.hpp" #include "Body.hpp" #include "RefVolume.hpp" +#include "Shell.hpp" #include "RefFace.hpp" #include "RefEdge.hpp" #include "RefVertex.hpp" @@ -57,7 +58,7 @@ PRINT_INFO("Operation Failed"); int ret_val = ( CubitMessage::instance()->error_count() ); - if ( ret_val != 7 ) + if ( ret_val != 0 ) { PRINT_ERROR("Errors found during Mergechk session.\n"); } @@ -119,5 +120,25 @@ gti->bodies(from_bodies); //Bug 247 test, when this bug get fixed, the comment should be removed. // status = gmti->imprint(from_bodies, new_bodies, CUBIT_FALSE); + + DLIList<Body*> bodies; + bodies.clean_out(); + gti->bodies(bodies); + //delete all entities + gti->delete_Body(bodies); + + DLIList<RefEntity*> free_entities; + gti->get_free_ref_entities(free_entities); + for(int i = 0; i < free_entities.size(); i++) + gti->delete_RefEntity(free_entities.get_and_step()); + + Body* body = gmti->brick(10, 10, 10); + DLIList<Shell*> shells; + body->shells(shells); + ShellSM* shell = shells.get()->get_shell_sm_ptr(); + OCCShell* occ_shell = CAST_TO(shell, OCCShell); + DLIList<TopologyBridge*> parents; + occ_shell->get_parents_virt(parents); + assert (parents.size() == 1); return status; }

1 0

r5289 - cgm/trunk/test
by jiangtao_ma＠yahoo.com 06 Jan '12

06 Jan '12

Author: janehu Date: 2012-01-06 09:26:29 -0600 (Fri, 06 Jan 2012) New Revision: 5289 Modified: cgm/trunk/test/makept.cpp cgm/trunk/test/modify.cpp Log: Changes make according to -r5288. Modified: cgm/trunk/test/makept.cpp =================================================================== --- cgm/trunk/test/makept.cpp 2012-01-06 15:23:12 UTC (rev 5288) +++ cgm/trunk/test/makept.cpp 2012-01-06 15:26:29 UTC (rev 5289) @@ -544,7 +544,11 @@ RefEdge* new_edge_1 = gmti->make_RefEdge(SPLINE_CURVE_TYPE, vertex1, vertex2, list); d = new_edge_1->measure(); - assert(fabs(d - 28.5)<0.01); + assert(fabs(d - 29.55)<0.01); + new_edge_1->closest_point_trimmed(center_pnt1, closest_location); + assert(center_pnt1.distance_between(closest_location) < 1.E-6); + new_edge_1->closest_point_trimmed(center_pnt2,closest_location); + assert(center_pnt2.distance_between(closest_location) < 1.E-6); //Spline with points and tangents CubitVector* first_v = new CubitVector(vertex1->coordinates()); Modified: cgm/trunk/test/modify.cpp =================================================================== --- cgm/trunk/test/modify.cpp 2012-01-06 15:23:12 UTC (rev 5288) +++ cgm/trunk/test/modify.cpp 2012-01-06 15:26:29 UTC (rev 5289) @@ -1266,8 +1266,8 @@ bodies.last(); body = bodies.get(); d = body->measure(); - //d = 93.697, no effect of draft angle. - assert(d - 93.697 < 0.001 && d > 93.697); + //d = 92.1335, no effect of draft angle. + assert(d - 92.1335 < 0.001 && d > 92.1335); bodies.clean_out(); gti->bodies(bodies);

1 0

Re: [cgma-dev] Create spline curve bug
by Jane Hu 06 Jan '12

06 Jan '12

Hi, Boyd This has been confirmed and fixed in most recent cgm. Thanks. Jane On Thu, Dec 22, 2011 at 11:23 AM, Boyd Tidwell <boyd(a)csimsoft.com> wrote: > Jane, > > The following: > > create vertex 0 0 0 > create vertex 1 0 1 > create vertex 2 0 0 > create curve vertex 1 2 3 spline > > should create a curve that includes all three vertices. It does so in > Cubit with the Acis engine but with OCC, the middle vertex (3) gets left > out. I looked a bit into the code with debug but I didn't come close to > having an idea what the problem may be. > > Thanks, > > - Boyd >

1 0

r5288 - cgm/trunk/geom/OCC
by jiangtao_ma＠yahoo.com 06 Jan '12

06 Jan '12

Author: janehu Date: 2012-01-06 09:23:12 -0600 (Fri, 06 Jan 2012) New Revision: 5288 Modified: cgm/trunk/geom/OCC/OCCModifyEngine.cpp Log: Fixed the issue brought up by Boyd Tidwell about make curve by points should pass all points instead of using those points as control points for B-Spline. Passed make check in cgm. Modified: cgm/trunk/geom/OCC/OCCModifyEngine.cpp =================================================================== --- cgm/trunk/geom/OCC/OCCModifyEngine.cpp 2012-01-05 21:41:38 UTC (rev 5287) +++ cgm/trunk/geom/OCC/OCCModifyEngine.cpp 2012-01-06 15:23:12 UTC (rev 5288) @@ -388,7 +388,7 @@ gp_Pnt pt; int size = 2+vector_list.size(); - TColgp_Array1OfPnt points(1, size); + Handle(TColgp_HArray1OfPnt) points = new TColgp_HArray1OfPnt(1, size); CubitVector vector; CubitVector closest_location; for(int i = 1; i <= size; i++) @@ -417,19 +417,22 @@ pt.SetCoord(closest_location.x(), closest_location.y(), closest_location.z()) ; } - points.SetValue(i, pt); + points->SetValue(i, pt); } //make curve according to the curve type. if(curve_type == SPLINE_CURVE_TYPE) { - Geom_BezierCurve BezierCurve(points); - Geom_BezierCurve* curve = new Geom_BezierCurve(BezierCurve); - Handle(Geom_BoundedCurve) curve_ptr(curve); - TopoDS_Vertex * vt1 = occ_point1->get_TopoDS_Vertex(); - TopoDS_Vertex * vt2 = occ_point2->get_TopoDS_Vertex(); - TopoDS_Edge new_edge = BRepBuilderAPI_MakeEdge(curve_ptr, *vt1, *vt2); - return OCCQueryEngine::instance()->populate_topology_bridge(new_edge, true); + GeomAPI_Interpolate spline(points, CUBIT_FALSE, TOL); + spline.Perform(); + if(spline.IsDone()) + { + Handle_Geom_BSplineCurve curve = spline.Curve(); + TopoDS_Vertex * vt1 = occ_point1->get_TopoDS_Vertex(); + TopoDS_Vertex * vt2 = occ_point2->get_TopoDS_Vertex(); + TopoDS_Edge new_edge = BRepBuilderAPI_MakeEdge(curve, *vt1, *vt2); + return OCCQueryEngine::instance()->populate_topology_bridge(new_edge, true); + } } return (Curve*) NULL;

1 0

OCC MacOS compile successful
by Tim Tautges 06 Jan '12

06 Jan '12

Hi all, Good news, last night my son James, hacker extraordinaire, got OCC 6.5.2 to compile successfully on MacOS. The details: - Snow leopard, gcc4.2, OCC6.5.2 (straight from the OCC download page) - Besides the following changes, follow the steps on http://opencascade.blogspot.com/2010/04/porting-on-macos.html - Run "autoreconf -fi" in the ros folder and then "aclocal", "automake", and "autoconf" - Add the following code to the platform specific part of ros/src/OSD/OSD_Chronometer.cxx: #ifdef __MACH__ # include <time.h> # include <sys/time.h> # include <mach/clock.h> # include <mach/mach.h> #endif - And then add this code to the OSD_Chronometer::GetThreadCPU function: #elif defined(__MACH__) struct timespec ts; clock_serv_t cclock; mach_timespec_t mts; host_get_clock_service(mach_host_self(), CALENDAR_CLOCK, &cclock); clock_get_time(cclock, &mts); mach_port_deallocate(mach_task_self(), cclock); ts.tv_sec = mts.tv_sec; ts.tv_nsec = mts.tv_nsec; - Proceed to configure, make, and install The new version of OSD_Chronometer.cxx is attached. We haven't built or tested CGM on top of this, but that's next. - tim -- ================================================================ "You will keep in perfect peace him whose mind is steadfast, because he trusts in you." Isaiah 26:3 Tim Tautges Argonne National Laboratory (tautges(a)mcs.anl.gov) (telecommuting from UW-Madison) phone (gvoice): (608) 354-1459 1500 Engineering Dr. fax: (608) 263-4499 Madison, WI 53706

1 0