Hi Wayne, This patch is not really useful as is, it was more of a test-bed for polygon tests and makes some invalid assumptions about what is expected from this code.
Seth has good ideas for more geometrically useful tests to do here. I plan to do something at some point, but not imminently, I'm afraid. Cheers, John On Thu, Jan 10, 2019 at 3:26 PM Wayne Stambaugh <[email protected]> wrote: > > Hi John, > > This patch no longer applies cleanly. Given the discussion, is this > something we still want to pursue? If so, please rebase this patch when > you get a chance. > > Cheers, > > Wayne > > On 12/19/2018 10:19 AM, John Beard wrote: > > Sorry, I failed to rebase that over the newest changes. This should > > apply better as a patch. > > > > Cheers, > > > > John > > On Wed, Dec 19, 2018 at 3:15 PM John Beard <[email protected]> wrote: > >> > >> Hi, > >> > >> I've added a few simple cases to the qa_common unit tests. > >> > >> I don't know if I will have covered cases which would have exposed any > >> bugs here, but any known corner cases (pass or fail) should be added. > >> > >> Also, they assume the polygons come in a fixed order. I don't know if > >> that's a correct assumption, or if the test should merely check the > >> triangle is somewhere in the list. > >> > >> These tests only cover PolygonTriangulation, not the use of it in > >> SHAPE_POLY_SET::CacheTriangulation(), so the Fracture bugs aren't > >> (yet...) covered. But it would probably be a good idea to also add a > >> test including those cases of SHAPE_POLY_SET and check that: > >> > >> 1) The tests fail before the recent fixes (i.e. the tests are > >> sensitive to the bug), and > >> 2) They do pass now > >> > >> Cheers, > >> > >> John > >> On Wed, Dec 19, 2018 at 1:57 PM jp charras <[email protected]> wrote: > >>> > >>> Le 19/12/2018 à 04:48, Seth Hillbrand a écrit : > >>>> Am 2018-12-18 14:19, schrieb jp charras: > >>>>> > >>>>> Sorry Seth, > >>>>> > >>>>> But with your fixes, CacheTriangulation() crashes with degenerated > >>>>> polygons. > >>>>> > >>>>> To see that, modify gerbview_painter.cpp, line 265 to remove > >>>>> absolutePolygon.COutline( 0 ).PointCount() < 3 > >>>>> and try to load the test file test_polygons_with_arcs.gbr in Gerbview. > >>>> > >>>> Hi JP- > >>>> > >>>> Thanks for finding that. The issue was that the Fracture() call > >>>> resulted in an empty polygon set with no outlines and I hadn't re-tested > >>>> its validity. 64f1fb9e7 fixes the bug. > >>>> > >>>> -Seth > >>> > >>> > >>> Looks good to me now. > >>> And CacheTriangulation() is really faster than GLU tesselation. > >>> Thanks. > >>> > >>> > >>> -- > >>> Jean-Pierre CHARRAS > >>> > >>> _______________________________________________ > >>> Mailing list: https://launchpad.net/~kicad-developers > >>> Post to : [email protected] > >>> Unsubscribe : https://launchpad.net/~kicad-developers > >>> More help : https://help.launchpad.net/ListHelp > >>> > >>> _______________________________________________ > >>> Mailing list: https://launchpad.net/~kicad-developers > >>> Post to : [email protected] > >>> Unsubscribe : https://launchpad.net/~kicad-developers > >>> More help : https://help.launchpad.net/ListHelp > > _______________________________________________ > Mailing list: https://launchpad.net/~kicad-developers > Post to : [email protected] > Unsubscribe : https://launchpad.net/~kicad-developers > More help : https://help.launchpad.net/ListHelp _______________________________________________ Mailing list: https://launchpad.net/~kicad-developers Post to : [email protected] Unsubscribe : https://launchpad.net/~kicad-developers More help : https://help.launchpad.net/ListHelp
