[cairo] [PATCH] Stroking bug (testcase & fix)

[Andrea Canciani]

I started studying the stroking code (since I'd like bug #24702 and
#17973 to be fixed) and I found a trivial bug when a dashes pattern of
odd length is used with a negative offset.
Regarding 24702, I think it could be solved by the "performance
improvement" (even if in this case it would become a real bug fix)
"Special-case extreme scaling of dashed lines..." in the TODO.
The idea proposed (replacing the dashing with a non-dashed transparent
color) looks very promising, but I think that this transformation
should not be done when every single dash is small, but only when the
total length of the pattern is small, otherwise a pattern like [1 0 1
0 1 ... 1] (odd length=n, thus it's almost the same as [n n] except
for caps) when scaled very small might become half-transparent too
early.
Andrea Canciani
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0001-test-Add-dash-offset-test.patch
Type: application/octet-stream
Size: 5565 bytes
Desc: not available
Url : http://lists.cairographics.org/archives/cairo/attachments/20091110/110515ba/attachment.obj 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 0002-Fix-odd-length-dashing-with-negative-offset.patch
Type: application/octet-stream
Size: 1225 bytes
Desc: not available
Url : http://lists.cairographics.org/archives/cairo/attachments/20091110/110515ba/attachment-0001.obj