Re: The rise and fall of Builders

2013-08-30 Thread Tom Eugelink


I know, the question was based on the best to cycle off builders remark, then 
what is advised to use as an alternative?

Tom


On 2013-08-30 06:56, Richard Bair wrote:

You can still use your own Builders and plug them into FXML. Its just that the 
built in set won't be there. But the Builder base class and all the FXML 
support is still there. All mentioned in that long thread :-D

Richard

On Aug 29, 2013, at 9:50 PM, Tom Eugelink t...@tbee.org wrote:


This week I ran into the problem that I needed to provide a date format 
(attribute in FXML) to one of my controls. So I needed a way to convert a 
string to DateFormat, or even a comma separated list to a list of DateFormats. 
This I solved with a builder for that control. How would one solved that 
without builders? (What is the alternative to builders?)

Tom


On 2013-08-29 22:42, Richard Bair wrote:

Deprecated in 8 and removed from the JavaDoc, gone in 9 (will be available as a 
separately downloadable Jar so you can keep using them, but they won't be 
updated). We're removing them from samples.

Best to cycle off the builders.

Richard

On Aug 29, 2013, at 12:42 PM, Felix Bembrick felix.bembr...@gmail.com wrote:


Thanks Jonathan,

So what is Oracle's current position on this?  Are Builders in or out?  If
out, when will they be removed and how?


On 30 August 2013 05:31, Jonathan Giles jonathan.gi...@oracle.com wrote:


You can catch up on the back story here:
http://mail.openjdk.java.net/pipermail/openjfx-dev/2013-March/006725.html

-- Jonathan

On 30/08/2013 7:06 a.m., Felix Bembrick wrote:

I was not privy to the original discussion but I am lead to believe that

Builders are no longer considered fashionable and that we are advised not
to use them.

While I realise that every type of Node basically needed its own

Builder, could someone please outline why this situation has arisen?  Is it
something to do with fluent APIs themselves or some other reason?

Thanks,

Felix







Re: The rise and fall of Builders

2013-08-30 Thread Richard Bair
Using your own builders is fine, I mean you best not use the deprecated ones as 
they will be disappearing.

On Aug 29, 2013, at 10:56 PM, Tom Eugelink t...@tbee.org wrote:

 
 I know, the question was based on the best to cycle off builders remark, 
 then what is advised to use as an alternative?
 
 Tom
 
 
 On 2013-08-30 06:56, Richard Bair wrote:
 You can still use your own Builders and plug them into FXML. Its just that 
 the built in set won't be there. But the Builder base class and all the FXML 
 support is still there. All mentioned in that long thread :-D
 
 Richard
 
 On Aug 29, 2013, at 9:50 PM, Tom Eugelink t...@tbee.org wrote:
 
 This week I ran into the problem that I needed to provide a date format 
 (attribute in FXML) to one of my controls. So I needed a way to convert a 
 string to DateFormat, or even a comma separated list to a list of 
 DateFormats. This I solved with a builder for that control. How would one 
 solved that without builders? (What is the alternative to builders?)
 
 Tom
 
 
 On 2013-08-29 22:42, Richard Bair wrote:
 Deprecated in 8 and removed from the JavaDoc, gone in 9 (will be available 
 as a separately downloadable Jar so you can keep using them, but they 
 won't be updated). We're removing them from samples.
 
 Best to cycle off the builders.
 
 Richard
 
 On Aug 29, 2013, at 12:42 PM, Felix Bembrick felix.bembr...@gmail.com 
 wrote:
 
 Thanks Jonathan,
 
 So what is Oracle's current position on this?  Are Builders in or out?  If
 out, when will they be removed and how?
 
 
 On 30 August 2013 05:31, Jonathan Giles jonathan.gi...@oracle.com wrote:
 
 You can catch up on the back story here:
 http://mail.openjdk.java.net/pipermail/openjfx-dev/2013-March/006725.html
 
 -- Jonathan
 
 On 30/08/2013 7:06 a.m., Felix Bembrick wrote:
 I was not privy to the original discussion but I am lead to believe that
 Builders are no longer considered fashionable and that we are advised not
 to use them.
 While I realise that every type of Node basically needed its own
 Builder, could someone please outline why this situation has arisen?  Is 
 it
 something to do with fluent APIs themselves or some other reason?
 Thanks,
 
 Felix
 
 
 



hg: openjfx/8/graphics/rt: RT-32603: Bad analysis in NGGroup#computeRenderRoot results in Nodes being rendered that need not be

2013-08-30 Thread hang . vo
Changeset: a2677d23fe0e
Author:rbair
Date:  2013-08-30 11:04 -0700
URL:   http://hg.openjdk.java.net/openjfx/8/graphics/rt/rev/a2677d23fe0e

RT-32603: Bad analysis in NGGroup#computeRenderRoot results in Nodes being 
rendered that need not be

! modules/graphics/src/main/java/com/sun/javafx/sg/prism/NGGroup.java
! modules/graphics/src/main/java/com/sun/javafx/sg/prism/NGNode.java
! modules/graphics/src/main/java/com/sun/javafx/sg/prism/NGRegion.java
! modules/graphics/src/test/java/com/sun/javafx/sg/prism/NGTestBase.java
! 
modules/graphics/src/test/java/com/sun/javafx/sg/prism/OcclusionCullingTest.java



Re: Bad quality of Node.snapshot

2013-08-30 Thread Richard Bair
Might be this one: https://javafx-jira.kenai.com/browse/RT-31909

On Aug 30, 2013, at 1:47 PM, Tom Schindl tom.schi...@bestsolution.at wrote:

 Hi,
 
 I've been using Node.snapshot() and found out that the quality of the
 snapshots is very bad (on OS-X Retina?).
 
 Take a look at http://downloads.efxclipse.org/screenshot.png the upper
 one is the really rendered element and the 2nd is a snapshot take from
 the control and rendered as an image.
 
 An idea why?
 
 Tom



Re: Bad quality of Node.snapshot

2013-08-30 Thread Richard Bair
Actually I was just on my way to fix this (and a couple other Region bugs) :-)

On Aug 30, 2013, at 2:07 PM, Richard Bair richard.b...@oracle.com wrote:

 Might be this one: https://javafx-jira.kenai.com/browse/RT-31909
 
 On Aug 30, 2013, at 1:47 PM, Tom Schindl tom.schi...@bestsolution.at wrote:
 
 Hi,
 
 I've been using Node.snapshot() and found out that the quality of the
 snapshots is very bad (on OS-X Retina?).
 
 Take a look at http://downloads.efxclipse.org/screenshot.png the upper
 one is the really rendered element and the 2nd is a snapshot take from
 the control and rendered as an image.
 
 An idea why?
 
 Tom
 



Re: Bad quality of Node.snapshot

2013-08-30 Thread Gerrit Grunwald
seems to be the same whe

Gerrit Grunwald

Westfalenstrasse 93
48165 Münster
- Germany -

mailhan.s...@muenster.de
webharmonic-code.org
twitter @hansolo_

Am 30.08.2013 um 22:47 schrieb Tom Schindl tom.schi...@bestsolution.at:

 Hi,
 
 I've been using Node.snapshot() and found out that the quality of the
 snapshots is very bad (on OS-X Retina?).
 
 Take a look at http://downloads.efxclipse.org/screenshot.png the upper
 one is the really rendered element and the 2nd is a snapshot take from
 the control and rendered as an image.
 
 An idea why?
 
 Tom


hg: openjfx/8/graphics/rt: RT-32639: NGRegion can avoid invalidating the backgroundInsets if the Background in question does not have any percentage-based CornerRadii on any BackgroundFills

2013-08-30 Thread hang . vo
Changeset: 49f80deda70a
Author:rbair
Date:  2013-08-30 14:04 -0700
URL:   http://hg.openjdk.java.net/openjfx/8/graphics/rt/rev/49f80deda70a

RT-32639: NGRegion can avoid invalidating the backgroundInsets if the 
Background in question does not have any percentage-based CornerRadii on any 
BackgroundFills

! modules/graphics/src/main/java/com/sun/javafx/sg/prism/NGRegion.java
! modules/graphics/src/main/java/javafx/scene/layout/Background.java
! modules/graphics/src/test/java/javafx/scene/layout/BackgroundTest.java



hg: openjfx/8/controls/rt: RT-31349: TextField, bidi text drifts when caret moves to bidi boundary.

2013-08-30 Thread hang . vo
Changeset: 089d0e8cb4d6
Author:leifs
Date:  2013-08-30 14:37 -0700
URL:   http://hg.openjdk.java.net/openjfx/8/controls/rt/rev/089d0e8cb4d6

RT-31349: TextField, bidi text drifts when caret moves to bidi boundary.

! 
modules/controls/src/main/java/com/sun/javafx/scene/control/skin/TextFieldSkin.java



Static FXMLLoader load method deprecation

2013-08-30 Thread John Smith
The static methods on FXMLLoader are very confusing.

It is very easy to create errors by writing code that mixes static and instance 
FXMLLoader methods.

Everything that can be done with static load methods, can be done with instance 
FXMLLoader methods.
Other than confusion, the static load methods seem to provide little value to 
me.

Is it possible to have the static FXMLLoader methods deprecated for Java 8?


Re: Static FXMLLoader load method deprecation

2013-08-30 Thread Daniel Zwolenski
+1



On 31/08/2013, at 8:54 AM, John Smith john_sm...@symantec.com wrote:

 The static methods on FXMLLoader are very confusing.
 
 It is very easy to create errors by writing code that mixes static and 
 instance FXMLLoader methods.
 
 Everything that can be done with static load methods, can be done with 
 instance FXMLLoader methods.
 Other than confusion, the static load methods seem to provide little value to 
 me.
 
 Is it possible to have the static FXMLLoader methods deprecated for Java 8?


hg: openjfx/8/graphics/rt: Minor comment fixes to NGRegion, also minor change to RegionImageCache to have the constructor take a ResourceFactory instead of a Graphics object (since the Graphics object

2013-08-30 Thread hang . vo
Changeset: a8e13d82e298
Author:rbair
Date:  2013-08-30 22:00 -0700
URL:   http://hg.openjdk.java.net/openjfx/8/graphics/rt/rev/a8e13d82e298

Minor comment fixes to NGRegion, also minor change to RegionImageCache to have 
the constructor take a ResourceFactory instead of a Graphics object (since the 
Graphics object was only used to get at its resource factory, and it is better 
style to pass what is really needed instead of a carrier).

! modules/graphics/src/main/java/com/sun/javafx/sg/prism/NGRegion.java
! modules/graphics/src/main/java/com/sun/javafx/sg/prism/RegionImageCache.java