Hi Phil,
I haven't looked too deep into that but to the "only" problem I see is
how the internal lookup tree is structured in "PrismFontFactory".
And yes you are right the OS/2-Table (there's even a comment in
PrismFontFile mentionning usWeightClass) contains the necessary
information, some mac-fonts use the header-segment so there one has to
distinction but those font would be simply classified as BOLD(700).
I'll put that on my opensource working queue, to look a bit deeper into
that and get back with a more in depth report (and probably questions).
The current in ability to use FontWeight is a real problem compared to
WebUIs, and although there is a work around not using FontWeight in CSS
it is very cumbersome.
Tom
Am 23.04.20 um 03:02 schrieb Philip Race:
I was mulling over why if it is so easy we didn't do it and I suspect
(haven't checked, and so long ago I can't remember off hand) that
this code needed to run on top of the old Java 2D pipeline (wrongly
called the Swing pipeline). And 2D did not and does not provide
this functionality either.
-phil.
On 4/22/20, 3:40 PM, Philip Race wrote:
Sadly no one is working on it.
It probably is not that hard to do, the weight should be in the OS/2
table
-phil
On 4/22/20, 3:26 PM, Tom Schindl wrote:
Hi Kevin / Phil,
I'd like to start working on JDK-8090423 but before I start investing
time I wanted to make sure nobody is working on that and if you have
any input, ideas already how to address that problem help me not
wasting time.
Tom
