Looks fine.
ps: for the record - long time ago the OSX code was different:
http://hg.openjdk.java.net/macosx-port/macosx-port/jdk/rev/e910dde2c36c
On 11.03.16 23:34, Phil Race wrote:
Oops. Fixed : http://cr.openjdk.java.net/~prr/8055463.1/
-phil.
On 03/11/2016 12:22 PM, Sergey Bylokhov wrote:
Hi, Phil.
The FileFont.java contains debug code:
System.out.println("COUNT="+count);
System.out.println("FILE="+fontFile);
On 11.03.16 2:02, Phil Race wrote:
PS there are a couple of other test files updated here
BigFont is one of the tests I ran whilst checking I didn't create
regressions
and it failed because it assumes a SecurityManager. I could not find an
open bug on that.*
*GetLCIDFromLocale happened to be the one I chose to paste in
the GPL and then I noticed it incorrectly had the classpath exception
so I am fixing that here too.
-phil.
On 03/10/2016 02:53 PM, Phil Race wrote:
https://bugs.openjdk.java.net/browse/JDK-8055463
http://cr.openjdk.java.net/~prr/8055463/
Proposal is to add two new methods in the Font class paralleling the
existing ones
that return only a single font.
Font[] createFont(File)
Font[] createFont(InputStream)
I tried to make the single & multiple code be shared as much as
possible
to avoid duplication and I eliminated code in the OSX sub-class which
seemed to be a pointless copy of the superclass.
The test checks the various assertions I intend by probing for
platform fonts.
-phil.
--
Best regards, Sergey.