Hi everyone,<br><br>here's a possible patch for the issue that has been discussed on the mailing list for the last few weeks. It makes sure that the fallback surface methods use the .x and .y offsets from _cairo_surface_get_extents correctly. The Win32 surface seems to be the only surface where those members are ever different from zero, so that's probably why it slipped through.<br>
<br>View: <a href="http://sprunge.us/KOZU?diff">http://sprunge.us/KOZU?diff</a><br>Raw: <a href="http://sprunge.us/KOZU">http://sprunge.us/KOZU</a><br><br>Please do comment.<br><br>Best regards<br>Ingmar Runge<br>