Questions

Revision: 4369 Author: gruehle@adobe.com Date: 2008-12-19 16:19:45 -0800 (Fri, 19 Dec 2008) Log Message: ----------- Update SkinTest for contentColor --> contentBackgroundColor change. Fixed compile errors in WheelLayout.as. Modified Paths: -------------- flex/sdk/trunk/development/eclipse/flex/flex4test/src/SkinTest.mxml flex/sdk/trunk/development/eclipse/flex/flex4test/src/layouts/WheelLayout.as

Revision: 4368 Author: gruehle@adobe.com Date: 2008-12-19 16:18:43 -0800 (Fri, 19 Dec 2008) Log Message: ----------- Style name change: contentColor --> contentBackgroundColor. QE Notes: Tests using contentColor need to be updated Doc Notes: None Bugs: --- Reviewer: --- Modified Paths: -------------- flex/sdk/trunk/frameworks/projects/flex4/defaults.css flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/ItemRenderer.as flex/sdk/trunk/frameworks/projects/flex4/src/mx/skins/spark/FxDefaultItemRenderer.mxml flex/sdk/trunk/frameworks/projects/flex4/src/mx/skins/spark/FxListSkin.mxml flex/sdk/trunk/frameworks/projects/flex4/src/mx/skins/spark/FxTextAreaSkin.mxml flex/sdk/trunk/frameworks/projects/flex4/src/mx/skins/spark/FxTextInputSkin.mxml flex/sdk/trunk/frameworks/projects/flex4/src/mx/skins/spark/SparkSkin.as flex/sdk/trunk/frameworks/projects/flex4/src/mx/styles/metadata/SparkColorStyles.as

Revision: 4336 Author: gruehle@adobe.com Date: 2008-12-16 17:07:11 -0800 (Tue, 16 Dec 2008) Log Message: ----------- Update the SkinTest.mxml file: - Add FxButtonBar - Add color theme selection buttons - Re-arrange the controls at the top QE Notes: None Doc Notes: None Bugs: None Reviewer: None Modified Paths: -------------- flex/sdk/trunk/development/eclipse/flex/flex4test/src/SkinTest.mxml

Revision: 4334 Author: gruehle@adobe.com Date: 2008-12-16 16:44:11 -0800 (Tue, 16 Dec 2008) Log Message: ----------- Removed DropShadowFilter from the ScrollBar and NumericStepper/Spinner button skins. These were added as part of the styling changes in revision 4314 but caused a large increase in memory usage. These have been replaced with gradient approximations. Added initial skin for ButtonBar. The corners of the first/last button skins get distorted when the buttons are resized. This will be fixed once scale grid support is done. Fixed style metadata name problem: highlightColor should be rollOverColor. QE Notes: Bitmap tests for scrollbar and numeric stepper/spinner need to be updated. Doc Notes: None Bugs: SDK-18452 Tests: checkintets Ticket Links: ------------ http://bugs.adobe.com/jira/browse/SDK-18452 Modified Paths: -------------- flex/sdk/trunk/frameworks/projects/flex4/src/Flex4Classes.as flex/sdk/trunk/frameworks/projects/flex4/src/mx/skins/spark/FxButtonBarSkin.m

Revision: 4314 Author: gruehle@adobe.com Date: 2008-12-15 14:40:02 -0800 (Mon, 15 Dec 2008) Log Message: ----------- Initial checkin of style support in the Spark skins. See the "Styling Gumbo Components" spec for details. QE Notes: Bitmap compares will be different for many skins. Joan will update the bitmaps. Doc Notes: See spec for details. Bugs: - Reviewer: Corey Modified Paths: -------------- flex/sdk/trunk/frameworks/mxml-2009-manifest.xml flex/sdk/trunk/frameworks/projects/flex4/defaults.css flex/sdk/trunk/frameworks/projects/flex4/manifest.xml flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/ItemRenderer.as flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/baseClasses/FxComponent.as flex/sdk/trunk/frameworks/projects/flex4/src/mx/components/baseClasses/GroupBase.as flex/sdk/trunk/frameworks/projects/flex4/src/mx/skins/spark/FxButtonSkin.mxml flex/sdk/trunk/frameworks/projects/flex4/src/mx/skins/spark/FxCheckBoxSkin.mxml flex/sdk/trunk/frameworks/p

Revision: 4313<br />Author: gauravj@adobe.com<br />Date: 2008-12-15 14:29:00 -0800 (Mon, 15 Dec 2008)<br /><br />Log Message:<br />-----------<br />adding new option to <asdoc> task.<br /><br />QE Notes: None<br />Doc Notes: None <br />tests: checkintests<br /><br />Modified Paths:<br />--------------<br /> flex/sdk/trunk/modules/antTasks/src/flex/ant/AsDocTask.java

Revision: 4312 Author: gauravj@adobe.com Date: 2008-12-15 14:01:12 -0800 (Mon, 15 Dec 2008) Log Message: ----------- using -date-in-footer=false will remove date from the footer. default is true. QE Notes: None Doc Notes: None Bugs: SDK-18236 tests: checkintests Ticket Links: ------------ http://bugs.adobe.com/jira/browse/SDK-18236 Modified Paths: -------------- flex/sdk/trunk/asdoc/templates/asdoc-util.xslt flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/asdoc/ASDocConfigHandler.java flex/sdk/trunk/modules/compiler/src/java/flex2/configuration_en.properties flex/sdk/trunk/modules/compiler/src/java/flex2/tools/ASDocConfiguration.java

Revision: 4298 Author: gauravj@adobe.com Date: 2008-12-12 12:27:08 -0800 (Fri, 12 Dec 2008) Log Message: ----------- ASDoc Bug fixes for: Inserting the [native] metadata tag between a class ASDoc comment and the class definition causes ASDoc to omit output for the class. Missing param description when using @9475319 tag. ASDoc comments on metadata tags other than Event, Effect, and Style should be applied to the next class element. Also making change to CompilerAPI.java so source from swc doesn't take precedence over source from disk. QE Notes: Some baselines need to updated. Doc Notes: None Bugs: SDK-18413, SDK-18415, SDK-18388 tests: checkintests, asdoc Ticket Links: ------------ http://bugs.adobe.com/jira/browse/SDK-18413 http://bugs.adobe.com/jira/browse/SDK-18415 http://bugs.adobe.com/jira/browse/SDK-18388 Modified Paths: -------------- flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/CompilerAPI.java flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/asdo

Hi All- I had just a general question regarding the private overLink property of the CursorManagerImpl. I have a use case where I need to determine what system cursor is being displayed when/if a user mouses over a particular component. For instance, if a user mouses over an anchor tag inside of a Text component's htmlText property, I would like to know if the hand cursor is being displayed at that moment. I have, so far, been unsuccessful in finding a way to do this in Flex 3 for Flash Player 9.0.124.0 (I have seen there are static properties available in Flash Player 10 but we're not yet targeting it for our users). My question is, why is the overLink a private member as opposed to protected or public? The only reason I could think of as to why this was done is for security reasons; more than likely surrounding XSS attacks etc. Here's a bug in the JIRA which is somewhat related to this: https://bugs.adobe.com/jira/browse/FP-652 The reason I want know which cursor is being display

Revision: 4268 Author: gauravj@adobe.com Date: 2008-12-09 10:55:35 -0800 (Tue, 09 Dec 2008) Log Message: ----------- Fix (again) asdoc for Mac. Modified Paths: -------------- flex/sdk/trunk/asdoc/build.xml

Revision: 4265 Author: gauravj@adobe.com Date: 2008-12-09 08:24:37 -0800 (Tue, 09 Dec 2008) Log Message: ----------- Fix asdoc for Mac. Modified Paths: -------------- flex/sdk/trunk/asdoc/build.xml

Revision: 4259 Author: gauravj@adobe.com Date: 2008-12-08 17:27:11 -0800 (Mon, 08 Dec 2008) Log Message: ----------- Fix for - Method names beginning in '$' cause an ASDoc error. QE Notes: None Doc Notes: None Bugs: SDK-18353 tests: checkintests Ticket Links: ------------ http://bugs.adobe.com/jira/browse/SDK-18353 Modified Paths: -------------- flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/asdoc/AsDocUtil.java

I am having difficulty in deciding where to store my FlexUnit tests. I am hoping that someone can provide some 'sage advice' for me.<br /><br />The project is simple. Here is the structure:<br /><br />\bin-debug<br />\src\impl\<several AS files><br />\src\unit_tests\\<several AS files><br />\src\main.mxml<br />\src\TestRunner.mxml<br /><br />'impl' is a package where all of my business logic is stored.<br />'unit_tests' are where all of my FlexUnit tests are.<br /><br />With this structure my unit tests cannot import and test any classes within the impl package unless I make the classes public instead of private. I do not want to do this since only 2 of the 10 classes within impl should be exposed by design. The others support those 2 top-level classes.<br /><br />I also want the ability to write unit tests for all of my classes, no matter where they reside. I also want to keep my

Sorry if this seems slightly off center, but I blame Adobe, not myself. Trying to determine the true status and positioning of AIR amongst the various Flex forums and projects for an outsider is practically impossible. So, I will be brief. In September Mr. Chambers sent opened the flow by telling us that we could find FP10 integrated with Flex SDK in something you call 1.5 or Cosmo, or both. He was properly circumspect, so I didn't rush into it -- there was not even an ADL binary when I checked. Now it is November, the Trunk is at 4005, and ostensibly, everything we need to start trying to take advantage of FP support of 3D operations is out there. But where? Where do we get a version of AIR 1.5 that can be installed on a target desktop, so that we can try to run the ADT test cycle? Call it Gumbo, call it Flex4, call it whatever, but can someone tell us how to start testing it for AIR application deployment? Thank you.

Revision: 4235 Author: gauravj@adobe.com Date: 2008-12-04 13:02:39 -0800 (Thu, 04 Dec 2008) Log Message: ----------- Fix for - Links and type information missing for air.update.event events in @7668748 listings Also adding support for multiple @7668748 for a method. QE Notes: None Doc Notes: None Bugs: SDK-18319 tests: checkintests Ticket Links: ------------ http://bugs.adobe.com/jira/browse/SDK-18319 Modified Paths: -------------- flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/asdoc/AsDocUtil.java flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/asdoc/ClassTable.java flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/asdoc/DocComment.java flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/asdoc/TopLevelClassesGenerator.java

Revision: 4227 Author: gauravj@adobe.com Date: 2008-12-03 15:44:14 -0800 (Wed, 03 Dec 2008) Log Message: ----------- Fix for - ASDoc is not displaying multiple examples correctly. QE Notes: None Doc Notes: None Bugs: SDK-18256 tests: checkintests Ticket Links: ------------ http://bugs.adobe.com/jira/browse/SDK-18256 Modified Paths: -------------- flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/asdoc/TopLevelClassesGenerator.java

Hi, I'm using path resolver (Application.setPathResolver) in one application to resolve some external resources. (Using flex compiler oem) But, tight now, this resources only exists as byte[] (retrieved from a remote server). In order to use current path resolve I need to save this data into a File. Is there any way to work with byte[] only? This is a sensible data and I prefer do not to save it at disk, even if is just for a moment. VELO

I have the following code in my test case: var token:AsyncToken = transfer.save(currentChild); token.addResponder(new AsyncResponder( addAsync( function (result:ResultEvent, token:Object=null):void { onSaveChild(result.result as Child); }, TIMEOUT), function(error:Object, token:Object=null):void {} ) ); When this is run, this gives me the following error in my debug: ArgumentError: Error #1063: Argument count mismatch on flexunit.framework::AsyncTestHelper/handleEvent(). Expected 1, got 2. at mx.rpc::AsyncResponder/result() at mx.rpc::AsyncToken/http://www.adobe.com/2006/flex/mx/internal::applyResult() at mx.rpc.events::ResultEvent/http://www.adobe.com/2006/flex/mx/internal::callTokenResponders() at mx.rpc::AbstractOperation/http://www.adobe.com/2006/flex/mx/internal::dispatchRpcEvent() at mx.rpc::AbstractInvoker/http://www.adobe.com/2006/flex/mx/internal::resultHandler() at mx.rpc::Responder/result() at mx.rpc::AsyncRequest/acknowledge() at NetConnectionMessageRe

When are we going to see some stability with the core flash.* class package and related code-hinting in Gumbo? The Matrix3D class is missing, the Vector class is being pulled in from the __AS3__ package (not good at all when moving code written in the Flash IDE over to Flex Builder, plus it is supposed to be a top-level class). One SDK build throws errors when trying to access the stage.nativeWindow property, another build fixes that but introduces other problems, there is no code-hinting for the Graphics class, the BitmapData fails to even show up in the list of classes the SDK can see... come on guys this needs to be sorted out. Getting the core classes stable really should be a priority at the moment considering most other libraries use or are built on the flash.* classes, additional eye-candy for the already bloated and slow Flex framework (mx.*) shouldn't be at the top of the to-do list.

Just want to make sure this isn't just my code.<br /><br />Using FlexUnit 0.9, on Flex 3.<br /><br />I have the following assertion in my TestCase:<br />assertEquals("Date values should match", date, currentBasic.date);<br /><br />Which returns the error:<br />Date values should match - expected:<Wed Dec 1 00:00:00 GMT+1000 1920> but was:<Wed Dec 1 00:00:00 GMT+1000 1920><br /><br />Which as you can see, is the same value.<br /><br />I have to change this to:<br />assertEquals("Date values should match", date.toString(), currentBasic.date.toString());<br /><br />for the assertion to pass.<br /><br />Is this expected behaviour or a bug in FlexUnit?

Hi, I'm getting a strange layout behavior in Gumbo. This seems to be a regression as I was not having the problem with the code of November 12th. Only after updating on November 21th, the problem arised. Here's a sample application: ------------------------------------------------------------------------ <?xml version="1.0" encoding="utf-8"?> <FxApplication xmlns=" http://ns.adobe.com/mxml/2009">     <layout>         <VerticalLayout />     </layout>         <VGroup>         <FxButton label="1" click="{t1.text='1'}" />         <TextGraphic id="t1" text="0" />         <FxButton label="2" click="{t1.text='2'}" />     </VGroup>         <HGroup>       &nbsp

Revision: 4184 Author: gauravj@adobe.com Date: 2008-11-25 12:33:43 -0800 (Tue, 25 Nov 2008) Log Message: ----------- Fix for @SEE tag referencing a non-global package function generates an empty link@href in DITA Also now the deprecated should show up in the property and method summary table. and when @SEE results in an invalid href, an additional attribute called invalidHRef will be added to the dita file - this isn't picked up by xslt but will be consumed by FB. QE Notes: Some baselines may need to be updated due to new attribute for invalidHRef Doc Notes: None Bugs: SDK-18206 tests: checkintests Ticket Links: ------------ http://bugs.adobe.com/jira/browse/SDK-18206 Modified Paths: -------------- flex/sdk/trunk/asdoc/templates/asdoc-util.xslt flex/sdk/trunk/asdoc/templates/class-files.xslt flex/sdk/trunk/modules/compiler/src/java/flex2/compiler/asdoc/TopLevelClassesGenerator.java

Hi, all, LiveDocs are a wonderful resource but here's a way to perhaps improve them. I recently wrestled with a problem and wondered if we can make life easier for those who do anything with itemRenderers. Many of us start with LiveDocs when developing with Flex and in my experience, there needs to be just a tad more about some of the internal workings of itemRenderers. I built a Flex app using an ADG which had two columns with itemRenderers. One in particular was painful to build because of the "recycling" issue. As far as I know, the LiveDocs don't mention anything about the fact that itemRenderers render a dozen or so rows at a time. Various articles on the Flex Developer Center (as well as many blogs) talked about this. But to a beginner, it won't be obvious why rows magically repeat ("recycle") values. Many hours will be spent wondering what is being done wrong when a little bit of heads-up in the LiveDocs could help users. If LiveDocs contain info on this issue already, maybe

Hey guys, Is it possible to get a hotfix or possible workaround for FP-780? I'd call it a show-stopper, I'm awfully fond of cut-n-paste 🙂 -Josh -- "Therefore, send not to know For whom the bell tolls. It tolls for thee." Like the cut of my jib? Check out my Flex blog! :: Josh 'G-Funk' McDonald :: 0437 221 380 :: josh@gfunk007.com :: http://flex.joshmcdonald.info/ :: http://twitter.com/sophistifunk

Removal of the Singleton design pattern has many benefits for Cairngorm with regards to Unit Testing and Modules. So far the list of Singletons used in Cairngorm are ServiceLocator, ModelLocator and CairngormEventDispacther. I've looked into solutions to replace all Singletons with Dependency Injection. So far here are a summary my conclusions. The ModelLocator. Singleton Purpose: A single point of reference for View or Commands to reference the Model. Solution: 1, Pass the model object via the event to the command. 2, Pass the model object to the view via reference (DI). All model instances are instantiated in a factory with all required dependencies passed via reference through the model object's constructor. The Model hierarchy represents the Display Object structure, therefore each presentation model should contain a reference to the next view's presentation model. ServiceLocator: Singleton Purpose: A single point of reference for Delegates to reference Services. Solution: