https://issues.apache.org/ooo/show_bug.cgi?id=125300
Armin Le Grand <armin.le.gr...@me.com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |ACCEPTED
Assignee|iss...@openoffice.apache.or |armin.le.gr...@me.com
|g |
Target Milestone|--- |4.2.0
Ever confirmed|0 |1
--- Comment #3 from Armin Le Grand <armin.le.gr...@me.com> ---
Problem happens in the print/PDF preparation when a metafile is prepared
(primitive stuff is okay). The metafile is clipped and the cropped part also
gets clipped. When these clip PolyPolygons are processed by the
MetaFileProcessor (and thus in logic coordinates) an already set VCL ClipRegion
collides with a 2nd VCL ClipRegion. To do this, these are clipped against each
other to AND them (basegfx::tools::clipPolyPolygonOnPolyPolygon).
In this special case the numerical nifties of double precision lead to an empty
result. The same works on pixel targets due to a different clipping mechanism.
It would be wrong to render the content when the result of clipping is empty;
this may be a valid result under various situations. But it is possible to
check both PolyPolygons being ranges (rectangles) and simplify execution. This
is anyways faster than the clipping and the default case.
Adding code to do this and checking...
--
You are receiving this mail because:
You are the assignee for the issue.
You are watching all issue changes.
