Re: [OpenJDK 2D-Dev] [9] RFR JDK-8164931 : Verify if writer.abort() works properly for all writers in IIOWriteProgressListener.

[Phil Race]

+1

-phil.

On 9/27/2016 8:26 AM, Brian Burkhalter wrote:

Hi Jay,

I think this looks better. +1.

Thanks,

Brian

On Sep 27, 2016, at 2:36 AM, Jayathirth D V <jayathirth....@oracle.com <mailto:jayathirth....@oracle.com>> wrote:

Hi Brain,
Thanks for your review.

I was keeping check for abortRequest() right after processImageProgess() call as I did for all other plugins. I have moved the complete processImageProgress() check to the end after last tile is written. Its better if we finish everything related to the write loop before we call processImageProgress().

Please find updated webrev for review:

http://cr.openjdk.java.net/~jdv/8164931/webrev.03/ <http://cr.openjdk.java.net/%7Ejdv/8164931/webrev.03/>

Thanks,
Jay
*From:*Brian Burkhalter
*Sent:*Tuesday, September 27, 2016 4:54 AM
*To:*Philip Race
*Cc:*Jayathirth D V; 2d-dev

*Subject:*Re: [OpenJDK 2D-Dev] [9] RFR JDK-8164931 : Verify if writer.abort() works properly for all writers in IIOWriteProgressListener.

My off-list question was this:

    I am probably missing something, but with respect to the TIFF
    writer, why is it necessary to move old lines 2607-2610 to new
    lines 2596-2599? In the old version, for whatever it’s worth, the
    tile offset and byte count are written before the abort, so if
    one wanted for some reason to read the output image up to the
    point of where the write is aborted the last tile would be
    accessible whereas in the new code it would not be.

Thanks,
Brian

On Sep 26, 2016, at 4:12 PM, Philip Race <philip.r...@oracle.com <mailto:philip.r...@oracle.com>> wrote:

    Seems OK .. modulo Brian's off-list question about TIFF