[svn:fx-4.x] 14492: Mirroring performance fixes

Revision: 14492
Revision: 14492
Author:   [email protected]
Date:     2010-03-01 17:15:48 -0800 (Mon, 01 Mar 2010)
Log Message:
Mirroring performance fixes
Fixed a performance problem in UIComponent. Updated invalidateLayoutDirection to only iterate through its non-style children (GraphicElements) if its layoutDirection has not changed.
http://bugs.adobe.com/jira/browse/SDK-25547 - Performance regression in datagrid scroll after 14208
Restored the use of transform.concatenatedMatrix instead of MatrixUtil.getConcatenatedMatrix. We want the postLayoutTransformOffsets because the Player incorporates these values in textWidth and textHeight.
Always return the absolute measuredWidth/Height
QE notes: none
Doc notes: none
Bugs: SDK-25547
Reviewer: Hans
Tests run: checkintests
Is noteworthy for integration: no
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-25547
    http://bugs.adobe.com/jira/browse/SDK-25547
Modified Paths:
    flex/sdk/branches/4.x/frameworks/projects/framework/src/mx/core/UIComponent.as
    flex/sdk/branches/4.x/frameworks/projects/framework/src/mx/core/UITextField.as

Similar Messages

[svn:fx-4.x] 14345: Mirroring bug fixes

Revision: 14345
Revision: 14345
Author:   [email protected]
Date:     2010-02-22 15:10:45 -0800 (Mon, 22 Feb 2010)
Log Message:
Mirroring bug fixes
http://bugs.adobe.com/jira/browse/SDK-25561 - MX PopUpButton displays popup inconsistently when layoutDirection is set to rtl
http://bugs.adobe.com/jira/browse/SDK-25560 - MX ColorPicker displays popup inconsistently when layoutDirection is set to rtl
Translate the position of the popup from the mirrored coordinates to the un-mirrored global coordinates
http://bugs.adobe.com/jira/browse/SDK-25552 - The Spark CheckBox icon should be mirroring
Removed setting the icon layoutDirection = ?\226?\128?\156ltr?\226?\128?\157. The icon should follow the mirroring of the component.
http://bugs.adobe.com/jira/browse/SDK-25484 - MX Slider datatip is incorrectly positioned when layoutDirection is rtl
Fixed two issues. The dataTip positioning is fixed by translating the position of the popup from the mirrored coordinates to the un-mirrored global coordinates. The second is fixing an issue of reversing the dataTipPosition when ?\226?\128?\156left?\226?\128?\157 or ?\226?\128?\156right?\226?\128?\157 when mirrored.
QE notes: None
Doc notes: None
Bugs: SDK-25561, SDK-25560, SDK-25552, SDK-25484
Reviewer: Hans
Tests run: checkintests
Is noteworthy for integration: No
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-25561
    http://bugs.adobe.com/jira/browse/SDK-25560
    http://bugs.adobe.com/jira/browse/SDK-25552
    http://bugs.adobe.com/jira/browse/SDK-25484
    http://bugs.adobe.com/jira/browse/SDK-25561
    http://bugs.adobe.com/jira/browse/SDK-25560
    http://bugs.adobe.com/jira/browse/SDK-25552
    http://bugs.adobe.com/jira/browse/SDK-25484
Modified Paths:
    flex/sdk/branches/4.x/frameworks/projects/framework/src/mx/controls/ColorPicker.as
    flex/sdk/branches/4.x/frameworks/projects/framework/src/mx/controls/PopUpButton.as
    flex/sdk/branches/4.x/frameworks/projects/framework/src/mx/controls/sliderClasses/Slider. as
    flex/sdk/branches/4.x/frameworks/projects/spark/src/spark/skins/spark/CheckBoxSkin.mxml

[svn:fx-4.x] 14376: Mirroring bug fixes

Revision: 14376
Revision: 14376
Author:   [email protected]
Date:     2010-02-23 16:31:35 -0800 (Tue, 23 Feb 2010)
Log Message:
Mirroring bug fixes
http://bugs.adobe.com/jira/browse/SDK-25427 - HTML AIR component should not be mirrored even when layoutDirection=rtl
Set layoutDirection=ltr in defaults.css
http://bugs.adobe.com/jira/browse/SDK-25626 - The Spark VSlider dataTip is not positioned correctly when layoutDirection=rtl. It obscures the thumb.
Translate the position from mirrored local coordinate to global coordinates.
http://bugs.adobe.com/jira/browse/SDK-25584 - MX MenuBar displays menus offscreen when layoutDirection is rtl
Added logic to keep the menus from going off the left side of the screen.
QE notes: none
Doc notes: none
Bugs: SDK-25427, SDK-25626, SDK-25584
Reviewer: Hans
Tests run: checkintests
Is noteworthy for integration: no
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-25427
    http://bugs.adobe.com/jira/browse/SDK-25626
    http://bugs.adobe.com/jira/browse/SDK-25584
    http://bugs.adobe.com/jira/browse/SDK-25427
    http://bugs.adobe.com/jira/browse/SDK-25626
    http://bugs.adobe.com/jira/browse/SDK-25584
Modified Paths:
    flex/sdk/branches/4.x/frameworks/projects/airframework/defaults.css
    flex/sdk/branches/4.x/frameworks/projects/framework/src/mx/controls/MenuBar.as
    flex/sdk/branches/4.x/frameworks/projects/spark/src/spark/components/VSlider.as

[svn:fx-4.x] 15006: Menu check icon mirroring bug fix

Revision: 15006
Revision: 15006
Author:   [email protected]
Date:     2010-03-24 16:38:11 -0700 (Wed, 24 Mar 2010)
Log Message:
Menu check icon mirroring bug fix
http://bugs.adobe.com/jira/browse/SDK-25976 - Un-mirror the check icon when layoutDirection=rtl for menu/menuBar
Update MenuCheck and MenuCheckDisabled to use layoutDirection=LTR
QE notes: none
Doc notes: none
Bugs: SDK-25976
Reviewer: Kevin
Tests run: checkintests
Is noteworthy for integration: no
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-25976
    http://bugs.adobe.com/jira/browse/SDK-25976
Modified Paths:
    flex/sdk/branches/4.x/frameworks/projects/sparkskins/src/mx/skins/spark/MenuCheck.mxml
    flex/sdk/branches/4.x/frameworks/projects/sparkskins/src/mx/skins/spark/MenuCheckDisabled .mxml
    flex/sdk/branches/4.x/frameworks/projects/wireframe/src/mx/skins/wireframe/MenuCheck.mxml
    flex/sdk/branches/4.x/frameworks/projects/wireframe/src/mx/skins/wireframe/MenuCheckDisab led.mxml

90% of browser rendering problems are caused by invalid code.   See details here:
http://validator.w3.org/check?verbose=1&uri=http%3A%2F%2Fwww.i-sensors.com%2F
You have a bunch of unclosed <img /> tags on your menus.
Nancy O.
Alt-Web Design & Publishing
Web | Graphics | Print | Media Specialists
http://alt-web.com/
http://twitter.com/altweb

MBP whine, 10.4.7, and the mirror widget fix

I am wondering about the 10.4.7 combo update for my MacBook Pro. It weighs in at 218 MB's, and as far as I can tell (from these forums) it contains the "keyboard update." My MacBook Pro is running 10.4.5, and the mirror widget "fix" fixes the whine.
I'd like to know if the 10.4.7 update disables the mirror widget fix. I'd also like to know if there is a way to uninstall the keyboard update after installing 10.4.7.
Also, what is Speedit? How do I install it? Is it something in Terminal? Does it fix the whine if 10.4.7 and the keyboard update is installed?
Thank you for your help!
Seth

The keyboard update is reported to kill the mirror widget hack. To restore it, see this thread:
http://discussions.apple.com/thread.jspa?threadID=540873&tstart=15
You don't really need the speedit.kext after the firmware update, you can use the "temperature monitor"; but if you still want to install it, do the following:
copy the .kext to your home folder, open terminal, type "sudo kextload -v speedit.kext".
It doesn't fix the whine as far as I know.
cheers...

[svn:fx-4.x] 15044: Mirroring Window Gripper and DateField bug fixes

Revision: 15044
Revision: 15044
Author:   [email protected]
Date:     2010-03-25 17:15:30 -0700 (Thu, 25 Mar 2010)
Log Message:
Mirroring Window Gripper and DateField bug fixes
http://bugs.adobe.com/jira/browse/SDK-25719 - The gripper does not show up in an MX Window that is mirrored and using flex chrome
Added an mx_internal property to Button to track if the Button should modify the layoutDirection of its skins. MX Window and MX WindowedApplication both set this property to undefined. This is the only skin/icon in the halo and spark framework that inherits its layoutDirection from its parent. All others have layoutDirection of LTR.
http://bugs.adobe.com/jira/browse/SDK-25995 - DateField opens drop down on wrong side when layoutDirection = rtl
Added check if left edge of the dropDown exceeds the left screen bounds.
QE notes: n/a
Doc notes: n/a
Bugs: SDK-25719, SDK-25995
Reviewer: Kevin
Tests run: checkintests, DateField
Is noteworthy for integration: no
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-25719
    http://bugs.adobe.com/jira/browse/SDK-25995
    http://bugs.adobe.com/jira/browse/SDK-25719
    http://bugs.adobe.com/jira/browse/SDK-25995
Modified Paths:
    flex/sdk/branches/4.x/frameworks/projects/airframework/src/mx/core/Window.as
    flex/sdk/branches/4.x/frameworks/projects/airframework/src/mx/core/WindowedApplication.as
    flex/sdk/branches/4.x/frameworks/projects/framework/src/mx/controls/Button.as
    flex/sdk/branches/4.x/frameworks/projects/framework/src/mx/controls/DateField.as

Originally posted by: hansm.science.uva.nlDELETE
In article <f7fd6k$kq5$[email protected]>,
"Igor Vinnykov" <[email protected]> wrote:
> We are pleased to announce a new release in the 1.1.x stream: version 1.1.3.
> Main target of 1.1.3 version is compatibility with Eclipse 3.3 and Mylyn.
> Starting from this version JavaHL becomes the default SVN client. This change
> caused by licensing topics and future project transfer to the eclipse.org:
> previous default client JavaSVN/SVNKit doesn't have a license, which allows
> defining it as default client, distributed from the eclipse.org.
>
> Important: Subversive is prepackaged only by JavaHL binaries for Windows. If
> you work under Linux, MacOS or other operation systems you should
> build/install required JavaHL binaries by yourself. JavaHL is a part of
> official Subversion source distribution available at
> http://subversion.tigris.org/project_packages.html.
I am sorry but this leaves me in the dark. On the site mentioned above I cannot
find something like a download for javaHL (on MacOSX especially).
Although I did install subversion 1.4.4 from the source distribution I do not
seem to have a javaHL binary. How further?
Hans van der Meer

[svn:fx-trunk] 5224: Flex SDK Bug Fixes

Revision: 5224
Author: [email protected]
Date: 2009-03-10 14:06:53 -0700 (Tue, 10 Mar 2009)
Log Message:
Flex SDK Bug Fixes
SDK-19772 Initially invisible graphic element can not be made visible
When visible is false, we were both setting the displayObject invisible and not drawing the graphics. When visible was then set to true, we changed the displayObject visibility, but didn't draw the graphics. The fix is to always draw the graphics when we are not sharing our displayObject. This way we simply toggle visibility on the displayObject on/off, instead of having to perform a redraw.
SDK-19581 BitmapFill incorrectly handle source property
Updated code to match code in BitmapGraphic. If the source is a class, we still need to check the type of the newly instantiated instance of that class.
SDK-19085 unexpected behaviour when setting scaleX < 3 on a LinearGradient
Internally, we no longer allow scaleX to be less than 2. Visually, this shouldn't have much impact.
QE Notes: n/a
Doc Notes: n/a
Bugs: SDK-19772, SDK-19581, SDK-19085
Reviewer: Evtim, Chet
Ticket Links:
http://bugs.adobe.com/jira/browse/SDK-19772
http://bugs.adobe.com/jira/browse/SDK-19581
http://bugs.adobe.com/jira/browse/SDK-19085
http://bugs.adobe.com/jira/browse/SDK-19772
http://bugs.adobe.com/jira/browse/SDK-19581
http://bugs.adobe.com/jira/browse/SDK-19085
Modified Paths:
flex/sdk/trunk/frameworks/projects/flex4/src/mx/graphics/BitmapGraphic.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/graphics/baseClasses/GraphicElement.as
flex/sdk/trunk/frameworks/projects/framework/src/mx/graphics/BitmapFill.as
flex/sdk/trunk/frameworks/projects/framework/src/mx/graphics/LinearGradient.as

error dateField not selecion date 27/11/2002 ?
This is bug flex 3 ?
thanks

Will hardwiring my mac mini to my router improve airplay mirroring performance?

I am currently running ATV 2 and a 2012 Mac Mini with Mountain Lion. The AirPlay Mirroring is working well but some videos are lagging and skipping.
If I hardwire my mac mini to the router will that improve performance?
Please advise and thank you for your help.

We've always been able to 'Airplay' from iTunes using wired or wifi (same for the older Airtunes I guess) - I think maybe the name is just a little confusing in that it implies wifi but that shouldn't be the only way of doing so.
There will be instances when you might want to mirror a wired Mac's display to AppleTV without physically connecting to the TV too.
As I say I just don't know if it's limited to wifi only for source mirroring devices as I don't have one capable of doing it.
AC

[svn] 4830: * Resubmitting 4795 with an additional fix for unresolved namespaces

Revision: 4830
Author:   [email protected]
Date:     2009-02-04 08:03:13 -0800 (Wed, 04 Feb 2009)
Log Message:
* Resubmitting 4795 with an additional fix for unresolved namespaces
that come from variables. For example:
    var ns:Namespace = new Namespace("blah");
    foo.ns::bar();
This is part of the fix for SDK-17798. It allows
DataBindingFirstPassEvaluator to find type information for
namespaceBT2 in:
But this isn't a complete fix, because I couldn't figure out how to
allow the PropertyWatcher for "test_namespace::namespaceBT2" to
access the property. The solution will probably be to provide the
open namespaces to the PropertyWatcher, but I haven't figured out
how to do this yet.
tests Passed: checkintests, performance testsuite
Needs QA: YES
Needs DOC: NO
Bug fixes: SDK-19047, SDK-17798 (partial fix)
API Change: NO
Reviewer: Erik
Code-level description of changes:
semantics/FlowAnalyzer.java
    See notes for revision 4795.
util/Context.java
    Modified processUnresolvedNamespaces() to report an error if we
    can't find a Slot for the namespace, but no longer report an error
    if we find a Slot without a Value. This happens when the
    namespace is defined in a variable. ConstantEvaluator will not
    have run yet, so the Slot for the namespace will not have a Value.
    We just silently punt in this case. The VM will report an error
    at runtime if the namespace can't be resolved. Unfortunately,
    data binding won't be able to track down change events in this
    case, though.
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-17798
    http://bugs.adobe.com/jira/browse/SDK-19047
    http://bugs.adobe.com/jira/browse/SDK-17798
Modified Paths:
    flex/sdk/trunk/modules/asc/src/java/macromedia/asc/semantics/FlowAnalyzer.java
    flex/sdk/trunk/modules/asc/src/java/macromedia/asc/util/Context.java

WordPress Performance - FIX

I installed WordPress 2.1 and was having SERIOUS performance issues, particularly in the admin screens where delays of upto 2 minutes were being experienced.
I had noticed that every now and again the URL in Safari would include the port number 16080, which puzzled me.
This port is used by the performance cache option in Apache, I turned this OFF and everything was blazing fast.
The option is in "Options" tab when editing a site using Server Admin.
Look here for more details:
<http://docs.info.apple.com/article.html?artnum=107898>
MacBook Pro Mac OS X (10.4.8)

While I have not set msg detail on to traack what would really happen, I don't think so. In reality, it is going to go through all the FY08 blocks to do the first calculation then the FY09 blocks to do that set of calculations. I think it would work like a loop within a loop. It finishes the fy08 loop before going on to the FY09 loop. By that time, most of the blocks would not be cached. Now if the Years were dense, it's a whole different story. I would prefer to use an if statement instead of the nested Fix statements and the blocks would be visited once.
I'm under the impression that Years is sparse( and all the other dimesions in the fix are as well), so no matter what, you are getting a specific number of blocks. Interestingly enough, It's possible the first way might be faster. I've seen instances where splitting a fix into two fixes allowed for a much smaller cache and was actually quickler.
You might want to run a test on sample basic with Set msg detail to see the order of blocks brought in for both the scenarions

[svn] 2312: CursorManager: New functionality and a fix for SDK-15029

Revision: 2312
Author: [email protected]
Date: 2008-06-30 17:24:01 -0700 (Mon, 30 Jun 2008)
Log Message:
CursorManager: New functionality and a fix for SDK-15029
* NEW: When the mouse leaves a window/stage where a cursor is set, the cursor no longer sits on the edge of the screen until the mouse comes back into the window, it disappears.
* NEW: In AIR, when the mouse leaves an active window and glides over an inactive window (without giving it focus) where a cursor is set, the system pointer will be hidden and the custom cursor will appear while gliding over that window, and disappear when you leave that window.
* FIX: When a cursor is set in a child Window, and the mouse moves over another Window or WindowedApplication, the cursor no longer disappears -- the correct system pointer or custom cursor will display.
* Fixed listener leaks when switching systemManagers (neither Jason nor I can tell when/why it would happen in CursorManager, however).
Reviewer: Jason
Bugs: SDK-15029
Doc: No
QA: Yes, here's an exhaustive list of ideas:
* We'll need tests for the new behaviors.
* For all tests, we need to try behavior in air AND behavior in (sa)player, and test regressions.
* Please test SDK-15583 again, all other cursor tests.
* Take the test case for SDK-15029 and open up TWO MyWindows. Test how the cursors change across all three app windows when they are overlapped like a Venn diagram, with different combinations of custom cursors on and off -- we should have test cases for all of this behavior.
* Test that custom cursors work properly when the mouse leaves a non-rectangular stage with transparency.
* Test mouse drag situations -- drag off stage, release, move back to stage; drag off and back onto stage, etc.
* In my tests, I found a lot of bad behavior in AIR, and player seemed to be fine. Do we have drag and drop use cases for AIR with custom cursors?
A list of some bugs I encountered:
* BUG (cosmetic): If the mouse is outside the stage when a cursor changes for the first time (say you toggle a checkbox which sets the cursor using the keyboard), the cursor will appear on the stage in the last known mouse position. When the mouse returns to the stage, everything will start working correctly again.
* BUG (cosmetic): In AIR, if you drag out, and back into the window, then release, a MOUSE_OUT is fired again over the stage, which is wrong. The mouse coords given are (-1,-1), also wrong. This causes the system cursor to show instead of the custom cursor.
* BUG: In AIR: if you drag off the edge of a Window with a custom cursor, the cursor will vanish. The MOUSE_OUT event correctly event fires while dragging off the window. Even though the handler tells the system mouse cursor to appear, it doesn't -- might be an AIR issue, since the mouse is still dragging when told to show. If you move back over the active window without dragging, the cursor goes back to normal.
* BUG: In AIR: if you drag over another window with a cursor, MOUSE_MOVE events don't fire on that other window until you let go. Therefore, a custom cursor in that window does not get drawn, nor does a system cursor. If you move back over the active window without dragging, the cursor goes back to normal.
* BUG: In AIR: When dragging, off a window, Event.MOUSE_LEAVE only fires after the mouse is let go (unlike MOUSE_OUT), EVEN IF the mouse is dragged back over the window and released over the window. So you're told the mouse left when it didn't.
* BUG: In Player: I could never get Event.MOUSE_LEAVE to fire when added to the stage in saplayer or browser.
Ticket Links:
http://bugs.adobe.com/jira/browse/SDK-15029
http://bugs.adobe.com/jira/browse/SDK-15029
http://bugs.adobe.com/jira/browse/SDK-15583
http://bugs.adobe.com/jira/browse/SDK-15029
Modified Paths:
flex/sdk/branches/3.0.x/frameworks/projects/framework/src/mx/managers/CursorManagerImpl.a s

How to perform fixed value mapping in XSLT?

Hi All,
Kindly guide me to do a XSLT mapping for fixed values (done in graphical mapping)
eg: if Band         Role
            A             JSE
            B               SE
and so on.
Kindly help me out to implement the above logic using xslt.
Thanks
Deepika

Hi,
Please find the links below.
xpath functions in xslt mapping
Design time Value-mappings in XSLT
File to Multiple IDocs (XSLT Mapping)
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/01a57f0b-0501-0010-3ca9-d2ea3bb983c1
https://www.sdn.sap.com/irj/servlet/prt/portal/prtroot/docs/library/uuid/9692eb84-0601-0010-5ca0-923b4fb8674a
Regards,
Phani

[svn:fx-4.x] 14275: Mirroring changes 2: Label rendering in PieSeries

Revision: 14275
Revision: 14275
Author:   [email protected]
Date:     2010-02-19 01:56:08 -0800 (Fri, 19 Feb 2010)
Log Message:
Mirroring changes 2: Label rendering in PieSeries
QE notes:
Doc notes:
Bugs:
Reviewer:
Tests run: checkintests - PASS
Is noteworthy for integration:
Modified Paths:
    flex/sdk/branches/4.x/frameworks/projects/datavisualization/src/mx/charts/series/PieSerie s.as

Daerandin wrote:
There is a chance you are being affected by the issue in this topic:
https://bbs.archlinux.org/viewtopic.php?id=180085
Try booting with the kernel parameter
rcutree.rcu_idle_gp_delay=1
Tried that, nothing changed

[svn:fx-4.x] 15312: ComboBox bug fix

Revision: 15312
Revision: 15312
Author:   [email protected]
Date:     2010-04-09 13:51:32 -0700 (Fri, 09 Apr 2010)
Log Message:
ComboBox bug fix
http://bugs.adobe.com/jira/browse/SDK-26089 - s:ComboBox is not updated correctly when selectedItem is removed from dataProvider
Handle this specific case by overriding itemRemoved and clearing the textInput. Changing updateLabelDisplay messes with the custom selected item logic, which is why I chose a more localized solution.
QE notes: none
Doc notes: none
Bugs: SDK-26089
Reviewer: Kevin
Tests run: ComboBox, checkintests
Is noteworthy for integration: no
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-26089
    http://bugs.adobe.com/jira/browse/SDK-26089
Modified Paths:
    flex/sdk/branches/4.x/frameworks/projects/spark/src/spark/components/ComboBox.as

Stope
Mailing me ***
Sent from my I phone

[svn:fx-4.x] 14696: Temporary FTETextField Fix

Revision: 14696
Revision: 14696
Author:   [email protected]
Date:     2010-03-10 19:08:30 -0800 (Wed, 10 Mar 2010)
Log Message:
Temporary FTETextField Fix
http://bugs.adobe.com/jira/browse/SDK-25669 - FTETextField relies on the render event
Temporary fix is to also listen for the ENTER_FRAME event when invalidating. A longer term solution will be investigated later.
QE notes: n/a
Doc notes: n/a
Bugs: SDK-25669
Reviewer: Alex
Tests run: checkintests
Is noteworthy for integration: yes
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-25669
    http://bugs.adobe.com/jira/browse/SDK-25669
Modified Paths:
    flex/sdk/branches/4.x/frameworks/projects/spark/src/mx/core/FTETextField.as

Stope
Mailing me ***
Sent from my I phone

[svn:fx-4.x] 14492: Mirroring performance fixes

Similar Messages

Maybe you are looking for