几何尺寸与公差论坛------致力于产品几何量公差标准GD&T (GDT:ASME)|New GPS(ISO)研究/CAD设计/CAM加工/CMM测量

几何尺寸与公差论坛------致力于产品几何量公差标准GD&T (GDT:ASME)|New GPS(ISO)研究/CAD设计/CAM加工/CMM测量 (http://www.dimcax.com/hust/index.php)
-   DirectDWG (http://www.dimcax.com/hust/forumdisplay.php?f=89)
-   -   【转帖】mirror extrusion problems with inserts and ellipitcal arcs (http://www.dimcax.com/hust/showthread.php?t=17211)

yang686526 2009-05-06 05:50 PM

【转帖】mirror extrusion problems with inserts and ellipitcal arcs
 
mirror / extrusion problems with inserts and ellipitcal arcs
mirror / extrusion problems with inserts and ellipitcal arcs
hi,
i have constructed a sample file which contains two inserts and two elliptical arcs. one of each has an extrusion vector 0/0/1, the other one has the vector 0/0/-1. in autocad i get the entities side by side (mirrored), but when i open the file with dwgdirect one lies upon the other.
is there a way to transform these entities so that they are displayed autocad like?
thanks,
torsten
attached files
i tried to draw the file in odamfcapp and i see the same results as in autocad.
what do you mean by "when i open the file with dwgdirect"?
sergey slezkin
hi sergey,
i am iterating through the modelspace, checking all entities.
to read the ellipse data, i do the following:
[...]
oddbentityptr odentityptr = m_pobjectiter->entity();
oddbentity *pentity = odentityptr.get();
oddbellipseptr odellipseptr = oddbellipse::cast(pentity);
if( oddbellipse *pellipse = odellipseptr.get() )
{
odgepoint3d odpt3d = pellipse->center();
double dr1 = len(pellipse->majoraxis().x, pellipse->majoraxis().y);
double dr2 = len(pellipse->minoraxis().x, pellipse->minoraxis().y);
double dphi = atan2( pellipse->majoraxis().y, pellipse->majoraxis().x );
double dang1 = pellipse->startangle();
double dang2 = pellipse->endangle();
}
[...]
it looks like the start and end angle are mixed up.
this is for the inserts:
[...]
oddbblockreferenceptr odblockreferenceptr = oddbblockreference::cast(pentity);
if( oddbblockreference *pblockreference = odblockreferenceptr.get() )
{
odgepoint3dodpt3d = pblockreference->position();
double dang = pblockreference->rotation();
odgescale3d odscale = pblockreference->scalefactors();
}
[...]
for both inserts i get the same insertion point although in the dxf file on of them has a negative x-value.
as far as i understand, if an element has an extrusion dwgdirect has calculated this already so i don't have to check the normal vector. but in this case it doesn't work.
thanks for your help.
torsten
for ellipses.
you code assumes that they are in xy plane.
angles are measured not around world's z axis from world's x axis but around normal counter-clock-wise from x axis of object's coordinate system (ocs). ocs is defined by normal according to autocad's arbitrary axis algorithm.
both ellipses have the same start and end angles. it's ok.
for inserts. in dxf file position is in ocs. position() function returns coordinates in wcs. in wcs position of both inserts is the same (try to select them in autocad and look at position in properties).
rotation of both inserts is the same - 0 but it's measured in different ocs.
sergey slezkin
hi sergey,
thanks for your quick response.
for the insert i am going to calculate the scaling according to its ocs - same is for the rotation of the elliptical arc.
torsten
just for your info:
oddbblockreference has methods
odgematrix3d blocktransform() and
odresult setblocktransform(const odgematrix3d&)
sometimes it's more convinient to use them instead of accessing position, scale, rotation and normal which being applied in correct order form the transformation matrix.
sergey slezkin
hi sergey,
thanks for your advice. i tested the function odgematrix3d blocktransform() but i couldn't find out how to get the scaling and rotation out of the matrix. could you give me some help?
thanks a lot,
torsten
if you need scaling and rotation separately you can use corresponding methods of oddbblockreference.
i wrote about blocktransform() function just because in some applications it's convinient to get transformation matrix instead of composing it manually from position, scale, rotation and normal.
sergey slezkin
ok - thanks for this information.
torsten


所有的时间均为北京时间。 现在的时间是 07:34 PM.