[prev in list] [next in list] [prev in thread] [next in thread] 

List:       openjdk-2d-dev
Subject:    Re: RFR: 8165943: LineBreakMeasurer does not measure correctly if TextAttribute.TRACKING is set.
From:       Phil Race <prr () openjdk ! org>
Date:       2022-11-29 22:58:06
Message-ID: POUnZIMW_TRbOyZxwEftPTyQbA_NvoLgLC-XHP5qpi0=.e6870ca7-67ed-4d9b-9b78-6e69362d6ffc () github ! com
[Download RAW message or body]

On Thu, 15 Sep 2022 16:51:27 GMT, Olga Mikhaltsova <omikhaltcova@openjdk.org> wrote:

> This is a fix for LineBreakMeasurer. It takes into account the \
> TextAttribute.TRACKING value (not eq 0) while calculating the line breaks. 
> Tested on Linux x64, Windows x64, macOS x64 with the reproducer \
> (LineBreakSample.java) attached to JDK-8165943 and the following group of tests: \
> `$JTREG_HOME/bin/jtreg -jdk:$BUILD_HOME ./test/jdk/java/awt/font`

I'm sure an automated test is possible - it should be easy enough to adjust the \
tracking and verify the breaks are different. The code paths here don't do complex \
text. Tracking doesn't really work for things like Arabic .. but it would be good to \
verify that Arabic measures as it lays out with tracking - ignoring it I expect, but \
I could be surprised if there's something I am overlooking. Hence why a test of that \
would be good.

-------------

PR: https://git.openjdk.org/jdk/pull/10289


[prev in list] [next in list] [prev in thread] [next in thread]