[OpenJDK Rasterizer] Marlin renderer contribution for review

[Laurent Bourgès]

Jim,

> There is one change that was pushed to the jdk9 master in the past couple
of weeks that we don't have yet in the GR-team repo.  I'm working on
synchronizing the two.  You can see the changelog here so you can get a
head start on making similar fixes on the Marlin renderer while I work on
the synch.

Do you mean I will have to submit another webrev based on the merged
forrest ?

> http://hg.openjdk.java.net/jdk9/client/jdk/rev/3850097b0fe9

Ok. As Andrea mentioned several times, Marlin does not have the fill bug
with custom composite.

I fixed it a long time ago when I reworked the clipping boundaries
(half-open intervals) and optimized the corresponding rendering
boundaries...

If it is the only change since march 2013, the submitted patch is up to
date.

I can still spend some time verifying clip / bounds again soon.

Clemens, are you listening ?
Could you give me your feedback on the webrev ?

As I said, I did not respect some coding convention, notably the max line
length = 80 in the Renderer class because it would make the code totally
unreadable due to the many loops / if blocks...

Andrea, could you confirm you agree contributing your work to OpenJDK ?

Dalibor, could you give your opinion on that process ?

Could anybody test the given Marlin patch with jtreg tests ? And j2dBench ?

Regards,
Laurent
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.openjdk.java.net/pipermail/graphics-rasterizer-dev/attachments/20150314/a1a77550/attachment.html>