> On Nov 15, 2015, at 02:29 AM, wolf <wv99999@gmail.com> wrote: > > […] calculate the area of the > triangle to be written, *after* its vertex vectors have been truncated to > final precision, and if the area is zero, do not write the triangle into the > .stl file. > This isn’t feasible as that would leave a hole in the mesh, making it non-manifold. Try it and see how various software complains about it. > […] Once code has been put together to allow > creating the cartesian coordinates of each vertex inside OpenSCAD, creating > 3D-offset functionality should be trivial, since all that is required is to > erect the normal onto each face, move the face by the required amount along > the normal, and reconnect each face to its neighbours. > This only works for convex objects. Concave objects would cause self-intersections which would have to be resolved. That’s the non-trivial aspects of such a function. I’m not sure what you’re trying to explain in the rest of your email (it’s getting a bit late here), so I’ll leave that so someone else :) -Marius

After some fine-tuning on the *fullnumstr()*, it seems to work nicely under more extensive tests. So at least for now, my lib has a working approach to show numbers a little more faithfully than built-in str() does: | fullnumstr ( n [d,dmax] )=str ( tested:60 ) | | Integer: | | 0> fullnumstr(0)= "0" | 1> fullnumstr(5)= "5" | 2> fullnumstr(12)= "12" | 3> fullnumstr(200)= "200" | 4> fullnumstr(1230)= "1230" | 5> fullnumstr(123456)= "123456" | 6> fullnumstr(1234567)= "1234567" | 7> fullnumstr(123456789)= "123456789" | 8> fullnumstr(123456000)= "123456000" | 9> fullnumstr(12345670)= "12345670" | 10> fullnumstr(123456789012345)= "123456789012345" | 11> fullnumstr(-123456789012345)= "-123456789012345" | 12> fullnumstr(12345678901234500)= "12345678901234500" | | Float: | | 13> fullnumstr(1234.5)= "1234.5" | 14> fullnumstr(12345.6)= "12345.6" | 15> fullnumstr(12345.67)= "12345.67" | 16> fullnumstr(0.12345)= "0.12345" | 17> fullnumstr(0.12)= "0.12" | 18> fullnumstr(0.123)= "0.123" | 19> fullnumstr(0.1234)= "0.1234" | 20> fullnumstr(0.123456)= "0.123456" | 21> fullnumstr(0.0000123456)= "0.0000123456" | 22> fullnumstr( 0.123456789012345)= "0.123456789012345" | 23> fullnumstr(-0.123456789012345)= "-0.123456789012345" | 24> fullnumstr(1/3)= "0.333333333333333" | 25> fullnumstr(2/3)= "0.666666666666667" | 26> fullnumstr(2/3,d=5)= "0.66667" | | Use d and dmax to set digits of fraction: | // Note the difference between d and dmax. | | 27> fullnumstr(5,d=2)= "5.00" | 28> fullnumstr(-5,d=2)= "-5.00" | 29> fullnumstr(5,dmax=2)= "5" | 30> fullnumstr(38,d=0)= "38" | 31> fullnumstr(38,d=2)= "38.00" | 32> fullnumstr(80,d=0)= "80" | 33> fullnumstr(3.80)= "3.8" | 34> fullnumstr(38.2,dmax=3)= "38.2" | 35> fullnumstr(38.2,d=3, dmax=2)= "38.20" | 36> fullnumstr(38.2,d=3, dmax=4)= "38.200" | 37> fullnumstr(38.2,d=5, dmax=4)= "38.2000" | 38> fullnumstr(3.839, d=0)= "4" | 39> fullnumstr(3.839, d=1)= "3.8" | 40> fullnumstr(3.839, d=2)= "3.84" | 41> fullnumstr(3.839, d=3)= "3.839" | 42> fullnumstr(3.839, d=4)= "3.8390" | 43> fullnumstr(3.839, d=1, dmax=2)= "3.8" | 44> fullnumstr(3.839, d=2, dmax=2)= "3.84" | 45> fullnumstr(3.839, d=3, dmax=2)= "3.84" | 46> fullnumstr(3.839, d=4, dmax=2)= "3.84" | 47> fullnumstr(43.839, d=4, dmax=2)= "43.84" | 48> fullnumstr(-3.839, d=1)= "-3.8" | 49> fullnumstr(-3.839, d=2)= "-3.84" | 50> fullnumstr(-3.839, d=4)= "-3.8390" | | 51> fullnumstr(0.123456789012345,d=6)= "0.123457" | 52> fullnumstr(0.123456789012345,d=7)= "0.1234568" | 53> fullnumstr(0.123456789012345,d=8)= "0.12345679" | 54> fullnumstr(0.123456789012345,d=9)= "0.123456789" | 55> fullnumstr(800.1,d=0)= "800" | 56> fullnumstr(800.01,d=0)= "800" | 57> fullnumstr(800.01,d=1)= "800.0" | 58> fullnumstr(800000000000.01,d=1)= "800000000000.0" | 59> fullnumstr(79.516556302228700,d=0)= "80" ----- $ Runsun Pan, PhD $ libs: doctest , faces ( git ), offline doc ( git ), runscad.py( 1 , 2 , git ); $ tips: hash( 1 , 2 ), sweep , var( 1 , 2 ), lerp , animGif , precision -- View this message in context: http://forum.openscad.org/Inconsistent-conversion-of-floating-number-to-string-at-7th-significant-digit-tp14350p14625.html Sent from the OpenSCAD mailing list archive at Nabble.com.

On 15 November 2015 at 15:51, Marius Kintel <marius@kintel.net> wrote: > > On Nov 15, 2015, at 02:29 AM, wolf <wv99999@gmail.com> wrote: > > > > […] calculate the area of the > > triangle to be written, *after* its vertex vectors have been truncated to > > final precision, and if the area is zero, do not write the triangle into > the > > .stl file. > > > This isn’t feasible as that would leave a hole in the mesh, making it > non-manifold. Try it and see how various software complains about it. > How would removing a zero-area triangle leave a non-zero-area hole?

You most likely end up with a vertex meeting an edge whereas vertices should always meet other vertices. The zero area triangle can have non zero length. On 31 December 2015 at 09:42, Triffid Hunter <triffid.hunter@gmail.com> wrote: > On 15 November 2015 at 15:51, Marius Kintel <marius@kintel.net> wrote: > >> > On Nov 15, 2015, at 02:29 AM, wolf <wv99999@gmail.com> wrote: >> > >> > […] calculate the area of the >> > triangle to be written, *after* its vertex vectors have been truncated >> to >> > final precision, and if the area is zero, do not write the triangle >> into the >> > .stl file. >> > >> This isn’t feasible as that would leave a hole in the mesh, making it >> non-manifold. Try it and see how various software complains about it. >> > > How would removing a zero-area triangle leave a non-zero-area hole? > > > _______________________________________________ > OpenSCAD mailing list > Discuss@lists.openscad.org > http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org > >

You do end up with a zero area hole, but it is still topologically a hole. On 31 December 2015 at 09:55, nop head <nop.head@gmail.com> wrote: > You most likely end up with a vertex meeting an edge whereas vertices > should always meet other vertices. The zero area triangle can have non zero > length. > > On 31 December 2015 at 09:42, Triffid Hunter <triffid.hunter@gmail.com> > wrote: > >> On 15 November 2015 at 15:51, Marius Kintel <marius@kintel.net> wrote: >> >>> > On Nov 15, 2015, at 02:29 AM, wolf <wv99999@gmail.com> wrote: >>> > >>> > […] calculate the area of the >>> > triangle to be written, *after* its vertex vectors have been truncated >>> to >>> > final precision, and if the area is zero, do not write the triangle >>> into the >>> > .stl file. >>> > >>> This isn’t feasible as that would leave a hole in the mesh, making it >>> non-manifold. Try it and see how various software complains about it. >>> >> >> How would removing a zero-area triangle leave a non-zero-area hole? >> >> >> _______________________________________________ >> OpenSCAD mailing list >> Discuss@lists.openscad.org >> http://lists.openscad.org/mailman/listinfo/discuss_lists.openscad.org >> >> >