[svn:fx-trunk] 10158: Reverting the styles-related change in commit 10131.

Revision: 10158
Author:   [email protected]
Date:     2009-09-11 09:47:12 -0700 (Fri, 11 Sep 2009)
Log Message:
Reverting the styles-related change in commit 10131. The fix broke several Mustella tests because of the change in checkbox/radiobutton padding. Verifying that this is the new look we want, also adjusting the code to make sure that we only do it for spark-skin cases.
Modified Paths:
    flex/sdk/trunk/frameworks/projects/framework/src/mx/controls/CheckBox.as
    flex/sdk/trunk/frameworks/projects/framework/src/mx/controls/RadioButton.as
    flex/sdk/trunk/tools/dependencychecker/frameworkSwcExceptionsList.txt

This is great. It will totally help clear up what commits happen in flex trunk vs the i10/11 branches.
Great solution!

Similar Messages

[svn:fx-trunk] 12912: With TLF 571 checked in we can now revert the MXML 2009 changes for how the id language attribute is processed .

Revision: 12912
Revision: 12912
Author:   [email protected]
Date:     2009-12-14 12:08:35 -0800 (Mon, 14 Dec 2009)
Log Message:
With TLF 571 checked in we can now revert the MXML 2009 changes for how the id language attribute is processed.
An unqualified id property will no longer be seen as a setter for a normal public, writable, non-static "id" property - which matches Flex 3 behavior.
As in earlier versions of Flex, the mechanism to retain MXML id attribute information at runtime is to implement mx.core.IMXMLObject. The compiler will generate a call to the intialized() method and pass in the id information.
Note: One exception to this rule seems to be UIComponent which instead get id attribute information via mx.core.IDeferredInstantiationUIComponent (although there seems to be a TODO in the codebase to remove this mechanism and thus should not be relied on by third parties).
QE notes: Please add tests for the TLF id property on flow elements (div, span, p, etc).
Doc notes: Please check that we have not documented
Bugs: N/A
Reviewer: Paul
Tests run: Checkintests, local test cases.
Is noteworthy for integration: Yes
Modified Paths:
    flex/sdk/trunk/frameworks/projects/framework/src/FrameworkClasses.as
    flex/sdk/trunk/frameworks/projects/framework/src/mx/core/UIComponent.as
    flex/sdk/trunk/frameworks/projects/spark/src/spark/core/SpriteVisualElement.as
    flex/sdk/trunk/frameworks/projects/spark/src/spark/primitives/supportClasses/GraphicEleme nt.as
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/mxml/builder/ComponentBuilder.jav a
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/mxml/lang/StandardDefs.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/mxml/reflect/Type.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/mxml/reflect/TypeTable.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/mxml/rep/init/ValueInitializer.ja va
Removed Paths:
    flex/sdk/trunk/frameworks/projects/framework/src/mx/core/IID.as

This is a duplicate post. This should be locked to avoid further confusion.
My CSS wont change
Nancy O.

[svn:bz-trunk] 11030: Tweak the deserialization of ASObjects to treat an empty string for the type of an object as null .

Revision: 11030
Author:   [email protected]
Date:     2009-10-20 11:35:02 -0700 (Tue, 20 Oct 2009)
Log Message:
Tweak the deserialization of ASObjects to treat an empty string for the type of an object as null. It appears that there is some logic in the LC remoting code that relies on a non-null class name to always exist. This change reverts to the old behavior of not allowing empty string as a value for the ASObject.namedType.
This should fix bug 2448442 and its duplicates caused by the recent serialization changes.
I don't think this is the perfect fix. Pending further investigation, a better fix would be either:
a. If it's OK to assume that empty string should always mean null for the type of the ASObject, the code that enforces it should be in the setter/getter inside ASObject and not in the deserializer.
b. ASObject doesn't guarantee that a named type exists or is valid. In that sense an empty string is as bad as some random characters that cannot be a valid class name in java, so depending on how disruptive it may be, the fix should be in any logic that uses ASObject.getType().
Modified Paths:
    blazeds/trunk/modules/core/src/flex/messaging/io/amf/AbstractAmfInput.java

Hi Pavan,
"In your payload there is no namespace prefix for the elements under PayloadHeader element."
Yes, you are right - but this message is standard AQ Adapter Header message - it's not defined by me. I just used message which was automatically added to my project when I have defined AQ Adapter.
"In your process is the default namespace is same as namespace value of tns ??"
Do you mean targetNamespace? If yes it's different as it points to process "targetNamespace="http://xmlns.oracle.com/PF_SOA_jws/PF_APPS/APPS_PROCESS" (names of application and process have changed as I try different ways to do that)
ns1 is: xmlns:ns1="http://xmlns.oracle.com/pcbpel/adapter/aq/PF_SOA/PF_APPS/PO_AQ"
"another thing is tns and ns1 should have same values.."
When I create a variable of header type, namespace ns1 is automatically created for it. I set it as property of receive activity. When process is instantiated on the serwer I get the error in which you can see that namespace is tns.
Maybe I'm doing something wrong but I don't see how I could fix this in my process.
You can see that the message I get on the server has nothing in common with the application/project/process names. Is it possible to define such variable?
Regards
Pawel
PS:
In Transformation xsl file, both variables (source and target) has tns namespace for Header and PayloadHeader, and no namespace for subfields.
Edited by: pawel.fidelus on 2010-01-05 02:37

[svn:bz-trunk] 19323: Revert fix for BLZ-578 (checkin 19214).

Revision: 19323
Revision: 19323
Author:   [email protected]
Date:     2010-12-13 12:10:28 -0800 (Mon, 13 Dec 2010)
Log Message:
Revert fix for BLZ-578 (checkin 19214). The BlazeDS/LCDS Spring integration code had a dependency on the thread local stuff that was removed as part of this bug fix. Revert the fix so we can move forward with lockdown testing. We can add this change back in when we resolve the dependency issue (tracked by Watson 2774331) if that's the appropriate thing to do.
Ticket Links:
    http://bugs.adobe.com/jira/browse/BLZ-578
Modified Paths:
    blazeds/trunk/modules/core/src/flex/messaging/FlexContext.java
    blazeds/trunk/modules/core/src/flex/messaging/MessageBroker.java
    blazeds/trunk/modules/core/src/flex/messaging/MessageBrokerServlet.java
    blazeds/trunk/modules/core/src/flex/messaging/MessageException.java

[svn:fx-trunk] 11575: Put default style declarations into one class per application or module.

Revision: 11575
Author:   [email protected]
Date:     2009-11-09 11:34:57 -0800 (Mon, 09 Nov 2009)
Log Message:
Put default style declarations into one class per application or module.
Generate all the default styles in one class instead of a class for each style. The name of the style class will be based on the application or module name. An application named ?\226?\128?\156foo?\226?\128?\157 will have a style class named ?\226?\128?\156_foo_Style?\226?\128?\157. The idea is to allow applications to be compiled with different themes and get their owns styles. Currently this is not possible because a global style in both themes will have the same class name, _globalStyle. Whatever class the top-level application loads the sub-applications and modules will have to use the same class because of the flash player first-class-in-wins rule. Now each application and module will have their own style class.
QE notes: None.
Doc notes: None.
Bugs: SDK-22454
Reviewer: Paul, Pete
Tests run: checkintests, all mustella tests.
Is noteworthy for integration: No.
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-22454
Modified Paths:
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/css/StyleDef.vm
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/css/StylesContainer.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/tools/PreLink.java

Revision: 11575
Author:   [email protected]
Date:     2009-11-09 11:34:57 -0800 (Mon, 09 Nov 2009)
Log Message:
Put default style declarations into one class per application or module.
Generate all the default styles in one class instead of a class for each style. The name of the style class will be based on the application or module name. An application named ?\226?\128?\156foo?\226?\128?\157 will have a style class named ?\226?\128?\156_foo_Style?\226?\128?\157. The idea is to allow applications to be compiled with different themes and get their owns styles. Currently this is not possible because a global style in both themes will have the same class name, _globalStyle. Whatever class the top-level application loads the sub-applications and modules will have to use the same class because of the flash player first-class-in-wins rule. Now each application and module will have their own style class.
QE notes: None.
Doc notes: None.
Bugs: SDK-22454
Reviewer: Paul, Pete
Tests run: checkintests, all mustella tests.
Is noteworthy for integration: No.
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-22454
Modified Paths:
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/css/StyleDef.vm
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/css/StylesContainer.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/tools/PreLink.java

[svn:fx-trunk] 14986: Merge mirroring tool tip change 14985 from 4. x to trunk.

Revision: 14986
Revision: 14986
Author:   [email protected]
Date:     2010-03-24 09:47:07 -0700 (Wed, 24 Mar 2010)
Log Message:
Merge mirroring tool tip change 14985 from 4.x to trunk.
QE notes:
Doc notes:
Bugs:
Reviewer:
Tests run: checkintests
Is noteworthy for integration:
Modified Paths:
    flex/sdk/trunk/frameworks/projects/framework/src/mx/managers/ToolTipManagerImpl.as
    flex/sdk/trunk/tools/dependencychecker/frameworkSwcExceptionsList.txt
Property Changed:
    flex/sdk/trunk/

[svn:fx-trunk] 10913: Reverted my previous reversion, since the original change wasn't the cause of the Mustella breakage.

Revision: 10913
Author:   [email protected]
Date:     2009-10-07 12:04:30 -0700 (Wed, 07 Oct 2009)
Log Message:
Reverted my previous reversion, since the original change wasn't the cause of the Mustella breakage. So once again:
The 'lineBreak' style is now properly declared as non-inheriting.
For a long time we've intended for it to be non-inheriting... it lives in BasicNonInheritingStyles.as! And we want it to be non-inheriting, because in TLF this format does not inherit from parent FlowElement to child FlowElement, so in Flex is should not inherit from parent UIComponent to child UIComponent. But the metadata incorrectly said inherit="yes". It now correctly says inherit="no".
QE notes: None
Doc notes: None
Bugs: SDK-23569
Reviewer: Ryan
Tests run: ant checkintests
Is noteworthy for integration: Yes
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-23569
Modified Paths:
    flex/sdk/trunk/frameworks/projects/spark/src/spark/styles/metadata/BasicNonInheritingText Styles.as

Due to a copy/paste glitch, some necessary spaces have inadvertently been removed. If I could fix this, I would.

[svn:fx-trunk] 9407: Reordering the if-statement in isMeasureFixed() exposed that hostFormat can be null when measure() is called if styles changed and measure is done before the next commitProperties .

Revision: 9407
Author:   [email protected]
Date:     2009-08-19 15:11:34 -0700 (Wed, 19 Aug 2009)
Log Message:
Reordering the if-statement in isMeasureFixed() exposed that hostFormat can be null when measure() is called if styles changed and measure is done before the next commitProperties. This states test exposed this.
We should rethink if we want to clear hostFormat rather than have a hostFormatChanged flag. If there is no hostFormat at measure then it has to be fixed because there is no line break format to check for auto-size.
QE notes:
Doc notes:
Bugs: SDK-22779
Reviewer: Gordon
Tests run: checkintests
Is noteworthy for integration: no
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-22779
Modified Paths:
    flex/sdk/trunk/frameworks/projects/spark/src/spark/primitives/RichEditableText.as

Oh my god, it is too long! You definitely check out types, casting and especially ODP.Net (it does everything for you)... etc. They can help you to simplify your code. I do not have enough time to copy paste it to Studio and understand and solve your issue, so I got title of your message as your main question.
In Oracle, you can create an autonumber field by using sequences object. This is really useful when you need to create a unique number to act as a primary key.
Basically you can create a sequence simply typing;
CREATE SEQUENCE MY_SEQUENCE;
now you have a sequence called "MY_SEQUENCE"... Then, I advice you select a number from sequence;
select MY_SEQUENCE.nextval from dual;
I said I advice actually kinda must, although it called sequence, I cannot be sequential. Do not even try to predict the value. You can be sure that it is unique number so you can use it.
Then insert you record and use that number part of your primary key. I think that's it. Have fun.

[svn:bz-trunk] 17772: Update the url in the tests to use a relative path instead of the default localhost : 8400 so that the tests can pass on appservers other than the default Tomcat .

Revision: 17772
Revision: 17772
Author:   [email protected]
Date:     2010-09-20 15:02:50 -0700 (Mon, 20 Sep 2010)
Log Message:
Update the url in the tests to use a relative path instead of the default localhost:8400 so that the tests can pass on appservers other than the default Tomcat.
Modified Paths:
    blazeds/trunk/qa/apps/qa-regress/testsuites/config/tests/EnforceEndpointValidation/Enforc eEndpointValidationFalseTest/Remoting_NetConnectionTest.mxml
    blazeds/trunk/qa/apps/qa-regress/testsuites/config/tests/EnforceEndpointValidation/Enforc eEndpointValidationTrueTest/Remoting_NetConnectionTest.mxml

Revision: 17772
Revision: 17772
Author:   [email protected]
Date:     2010-09-20 15:02:50 -0700 (Mon, 20 Sep 2010)
Log Message:
Update the url in the tests to use a relative path instead of the default localhost:8400 so that the tests can pass on appservers other than the default Tomcat.
Modified Paths:
    blazeds/trunk/qa/apps/qa-regress/testsuites/config/tests/EnforceEndpointValidation/Enforc eEndpointValidationFalseTest/Remoting_NetConnectionTest.mxml
    blazeds/trunk/qa/apps/qa-regress/testsuites/config/tests/EnforceEndpointValidation/Enforc eEndpointValidationTrueTest/Remoting_NetConnectionTest.mxml

[svn:fx-trunk] 13157: -revert a portion of my previous change so that asdoc will work properly - moved the proprietary automation swcs back into the libs directory .

Revision: 13157
Revision: 13157
Author:   [email protected]
Date:     2009-12-22 10:15:18 -0800 (Tue, 22 Dec 2009)
Log Message:
-revert a portion of my previous change so that asdoc will work properly - moved the proprietary automation swcs back into the libs directory. the automation projects swcs will still get compiled into the automation directory.
This should fix the build
QE notes: make sure packaging is correct
Doc notes: no
Bugs: no
Reviewer:no
Tests run: checkintests, doc targets run
Is noteworthy for integration: no
Modified Paths:
    flex/sdk/trunk/build.xml
    flex/sdk/trunk/frameworks/projects/automation/build.xml
    flex/sdk/trunk/frameworks/projects/automation_air/build.xml
    flex/sdk/trunk/frameworks/projects/automation_dmv/build.xml
    flex/sdk/trunk/frameworks/projects/automation_flashflexkit/build.xml
    flex/sdk/trunk/frameworks/projects/automation_spark/build.xml
Added Paths:
    flex/sdk/trunk/frameworks/libs/automation_agent.swc
    flex/sdk/trunk/frameworks/libs/qtp.swc
    flex/sdk/trunk/frameworks/libs/qtp_air.swc
Removed Paths:
    flex/sdk/trunk/frameworks/libs/automation/

Revision: 13157
Revision: 13157
Author:   [email protected]
Date:     2009-12-22 10:15:18 -0800 (Tue, 22 Dec 2009)
Log Message:
-revert a portion of my previous change so that asdoc will work properly - moved the proprietary automation swcs back into the libs directory. the automation projects swcs will still get compiled into the automation directory.
This should fix the build
QE notes: make sure packaging is correct
Doc notes: no
Bugs: no
Reviewer:no
Tests run: checkintests, doc targets run
Is noteworthy for integration: no
Modified Paths:
    flex/sdk/trunk/build.xml
    flex/sdk/trunk/frameworks/projects/automation/build.xml
    flex/sdk/trunk/frameworks/projects/automation_air/build.xml
    flex/sdk/trunk/frameworks/projects/automation_dmv/build.xml
    flex/sdk/trunk/frameworks/projects/automation_flashflexkit/build.xml
    flex/sdk/trunk/frameworks/projects/automation_spark/build.xml
Added Paths:
    flex/sdk/trunk/frameworks/libs/automation_agent.swc
    flex/sdk/trunk/frameworks/libs/qtp.swc
    flex/sdk/trunk/frameworks/libs/qtp_air.swc
Removed Paths:
    flex/sdk/trunk/frameworks/libs/automation/

[svn:fx-trunk] 10578: Fix the SkinTest.mxml in the sparkTest project: use backgroundColor style as we don't have a property anymore.

Revision: 10578
Author:   [email protected]
Date:     2009-09-24 15:02:04 -0700 (Thu, 24 Sep 2009)
Log Message:
Fix the SkinTest.mxml in the sparkTest project: use backgroundColor style as we don't have a property anymore.
QE notes: None
Doc notes: None
Bugs: None
Reviewer: Evtim
Tests run: build
Is noteworthy for integration: No
Modified Paths:
    flex/sdk/trunk/development/eclipse/flex/sparkTest/src/SkinTest.mxml

I finally found how to set a default skin to a custom component so that users of my component will not have to set again the skinClass value of my component.
You have to create a defaults.css file at the root of your library and tell the compiler to take it into account.
The remaining problem is about the compiler. There are some steps before success and they are a bit mysterious/unclear....
I found 2 or 3 blog articles explaining those steps but the compiler arguments to use are all differents in each article...
Here are the links I found:
http://www.unitedmindset.com/jonbcampos/2010/05/12/creating-custom-spark-components/
http://www.betadesigns.co.uk/Blog/2010/05/14/default-skin-for-custom-flashbuilder-componen ts/
http://flexdevtips.blogspot.com/2009/06/default-stylesheet-in-swc-flex-library.html
Following the first article guidelines has been successful for me but I'm not marking this topic as Resolved because I'd like some answers about this whole thing...

[svn:fx-trunk] 4960: Decreased the number of Gumbo style buckets.

Revision: 4960
Author: [email protected]
Date: 2009-02-13 18:23:27 -0800 (Fri, 13 Feb 2009)
Log Message:
Decreased the number of Gumbo style buckets.
Combined AdvancedCharacterFormatTextStyles.as, AdvancedParagraphFormatTextStyles.as, and AdvancedContainerFormatTextStyles.as into AdvancedTextLayoutFormatStyles.as.
Combined BasicCharacterFormatTextStyles.as, BasicParagraphFormatTextStyles.as, and BasicContainerFormatTextStyles.as into BasicTextLayoutFormatStyles.as.
The reason for doing this is that TLF combined its CharacterFormat, ParagraphFormat, and ContainerFormat classes into a single TextLayoutFormat.
QE Notes: None
Doc Notes: None
Bugs: None
Reviewer: None; no functional change
Modified Paths:
flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/FxButton.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/FxContainer.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/FxDataContainer.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/FxNumericStepper.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/FxScroller.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/TextView.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/baseClasses/FxSlider.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/baseClasses/FxTextBase.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/baseClasses/GroupBase.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/graphics/TextBox.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/graphics/TextGraphic.as
Added Paths:
flex/sdk/trunk/frameworks/projects/flex4/src/mx/styles/metadata/AdvancedTextLayoutFormatS tyles.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/styles/metadata/BasicTextLayoutFormatStyl es.as
Removed Paths:
flex/sdk/trunk/frameworks/projects/flex4/src/mx/styles/metadata/AdvancedCharacterFormatTe xtStyles.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/styles/metadata/AdvancedContainerFormatTe xtStyles.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/styles/metadata/AdvancedParagraphFormatTe xtStyles.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/styles/metadata/BasicCharacterFormatTextS tyles.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/styles/metadata/BasicContainerFormatTextS tyles.as
flex/sdk/trunk/frameworks/projects/flex4/src/mx/styles/metadata/BasicParagraphFormatTextS tyles.as

Gordon, it looks like its been a while since you made this post. Not sure how valid it is now... I am particularly interested in the LigatureLevel.NONE value. It seems that it is no longer supported.
How do I turn of ligatures in the font rendering?
My flex project involves trying to match the font rendering of Apache's Batik rendering of SVG and ligatures have been turned off in that codebase. Is there any way (even roundabout) to turn ligatures off in flash?
Thanks,
Om

[svn:fx-trunk] 10607: Moved the 'breakOpportunity' style from BasicInheritingTextStyles.as to AdvancedInheritingTextStyles.as , so it is declared on fewer components now.

Revision: 10607
Author:   [email protected]
Date:     2009-09-25 16:08:48 -0700 (Fri, 25 Sep 2009)
Log Message:
Moved the 'breakOpportunity' style from BasicInheritingTextStyles.as to AdvancedInheritingTextStyles.as, so it is declared on fewer components now. As far as I can tell, it is mainly useful in rich text.
QE notes: None
Doc notes: None
Bugs: SDK-22474
Reviewer: Ryan
Tests run: ant checkintests
Is noteworthy for integration: No
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-22474
Modified Paths:
    flex/sdk/trunk/frameworks/projects/spark/src/spark/components/Label.as
    flex/sdk/trunk/frameworks/projects/spark/src/spark/styles/metadata/AdvancedInheritingText Styles.as
    flex/sdk/trunk/frameworks/projects/spark/src/spark/styles/metadata/BasicInheritingTextSty les.as

Gordon, it looks like its been a while since you made this post. Not sure how valid it is now...   I am particularly interested in the LigatureLevel.NONE value. It seems that it is no longer supported.
How do I turn of ligatures in the font rendering?
My flex project involves trying to match the font rendering of Apache's Batik rendering of SVG and ligatures have been turned off in that codebase. Is there any way (even roundabout) to turn ligatures off in flash?
Thanks,
Om

[svn:fx-trunk] 8742: * Revert revisions 8682 and 8734 due to performance issues.

Revision: 8742
Author:   [email protected]
Date:     2009-07-22 14:07:24 -0700 (Wed, 22 Jul 2009)
Log Message:
Revert revisions 8682 and 8734 due to performance issues. I'll
recommit when they are worked out.
tests Passed: checkintests
Needs QA: YES
Needs DOC: NO
Bug unfixed: SDK-14251
API Change: NO
Is noteworthy for integration: No
Ticket Links:
    http://bugs.adobe.com/jira/browse/SDK-14251
Modified Paths:
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/CompilerAPI.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/SymbolTable.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/abc/AbcCompiler.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/abc/Method.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/abc/Variable.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/as3/As3Compiler.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/as3/binding/DataBindingFirstPassE valuator.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/as3/binding/TypeAnalyzer.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/as3/reflect/As3Class.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/as3/reflect/Method.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/as3/reflect/TypeTable.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/as3/reflect/Variable.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/mxml/reflect/TypeTable.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/mxml/rep/MxmlDocument.java
Removed Paths:
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/abc/Assignable.java
    flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/as3/reflect/Assignable.java

ould wrote:
I have done the hard work for you. My HTPC is pretty barebones so it was relatively easy to roll back all the packages that depend on libjpeg>7 with older versions. I then compiled the xbmc-svn package using that system. Slow downs are gone and I have uploaded the package to hotfile here:
http://hotfile.com/dl/13546051/021f9c9/ … ar.gz.html
This is for 32-bit. You will need the libjpeg6 package from AUR installed. I have tested it on two systems, my main system which has both libjpeg7 and the AUR libjpeg6 installed and on my htpc which has only the libjpeg6 along with the rolled back dependancies. Works good on both systems.
Hope that helps!
Kevin
Thanks ould!! You got it working with libjpeg6 again!
I had to upgrade libcdio again so now I lost dvd support. It's not easy to use the svn version.
If you could compile the build against libcdio 0.80 I'd be forever grateful... If you've got the time.
Last edited by Perre (2009-09-26 17:09:04)

[svn:fx-trunk] 9174: revert rev# 9148 until some odd behavior can be tracked down.

Revision: 9174
Author:   [email protected]
Date:     2009-08-07 10:37:23 -0700 (Fri, 07 Aug 2009)
Log Message:
revert rev# 9148 until some odd behavior can be tracked down.
checkintests: pass
rev# 9148:Changing enabled on GroupBase to set the alpha to the new "disabledAlpha" CSS style (0.5 by default) when the container is disabled.
Modified Paths:
    flex/sdk/trunk/frameworks/projects/spark/defaults.css
    flex/sdk/trunk/frameworks/projects/spark/src/spark/components/supportClasses/GroupBase.as

[svn:fx-trunk] 10158: Reverting the styles-related change in commit 10131.

Similar Messages

Maybe you are looking for