Recommended way to unit test mobile cross platform C++ in Visual Studio 2015

Hello,
I'm playing with the new Visual Studio 2015 to check its capabilities for cross platform development using C++ as coding language. I've seen several videos in Channel9 about this feature but I don't remember hearing in any of these a word about how
to unit test such projects from Visual Studio 2015. Is there some recommended way to create unit tests for such projects?
I'm not sure I can use Visual Studio test projects for C++ to test the android library or the Windows Phone 8.1. If I add a new platform ARM to a test project of those, I cannot select in the platform toolset combo the values that I have for my libraries
(Clang 3.4 and Windows 8.1 respectively). So it seems that such test projects are not addressed for cross platform c++ in VS2015.
I also have tried to make a second cross platform library project to contain the tests, adding there the Catch.h file from catch framework or adding the files from Google Test framework. My idea was to link each cross platform test library with their corresponding
"production code" libraries. In both cases (catch and gtest) I have problems when trying to compile the test framework files (not sure if it is not possible or just my fault being new at C++ for mobile platforms).
Has anyone succeeded unit testing the cross platform c++ code in VS2015? How?
Thanks in advance,
Jorge.

Hi Jorge,
Thanks you for posting in MSDN forum.
Based on your issue, since VS2015 Preview is previews for the next major release of Visual Studio, currently we forums don’t provide support for VS2015 Preview. So the product team would accept the suggestions or feedbacks for the VS2015 now.
So you can submit the feedback/suggestions to Microsoft Connect feedback portal:
http://connect.microsoft.com/VisualStudio/feedback/CreateFeedback.aspx, Microsoft engineers will evaluate them seriously. Thanks for your understanding.
After you submit the feedback, you can post the link here which will be beneficial for other members with the similar issue. And I will help you to vote it.
Thank you for your understanding.
Best Regards,
We are trying to better understand customer views on social support experience, so your participation in this interview project would be greatly appreciated if you have time. Thanks for helping make community forums a great place.
Click
HERE to participate the survey.

Similar Messages

  • Best way to Unit test Modules?

    I am working on a Project where the main app is a separate Flex project and then there are several Modules each having their own unit tests. What is the best recommended way of setting up something like this in terms of test runners? I would like to have one test runner that loads all others?

    I have never unit tested just a controller. I have always done end to end testing with the systems I have worked on, so I am coming from a different viewpoint than yourself. I would tend to think that just running a full unit test all the way to your database would be the best test. You will be more likely to find the issues with your program if you are testing with an environment as similar to production as possible.
    After running a test you can either purge the data you entered from your test by having sql scripts run against the db, or if this is difficult just reload the database from its original backup. You would probably be able to purge the data quicker than reloading the database from my experience, but it depends on your complexity of the sql scripts.
    best of luck.

  • Best way to Unit Test Controller Layer

    Hi, I was hoping to get the opinion on this matter. My group is currently discussing which method would be the best to Unit test the Controller Layer (using StrutsTestCase). Basically with StrutsTestCase, you can test your Action methods and verify that the action forwards to the correct place, or that certain errors come up. The problem is that these Action classes will make Model layer calls that affect the DB. Whats the best way to test the Action classes? Some possibilities are:
    #1) Make a mock model layer that instead of hitting the DB uses a "switch" statement to return the correct value.
    #2) Use a test DB to test this. The problem with this is that after running tests, you want to leave the test DB in the same state as when you started the test. This is not very easy to do if your web application only does inserts to the DB. In addition, the Model layer Unit testing will be designed such that it will insert, select, update, select, and delete (in that order), so the state of the DB will remain the same after the test is run (unless JUnit crashes halfway).
    #3) Use a copy of a test DB on your local computer. This is beneficial because you can do anything to the DB, and then just delete it because a "master" copy is located elsewhere. The problem with this is, we also hoped to do performance testing of the model layer by running multiple threads of the Model layer unit tests. This won't work now because each thread would make a copy of the master DB....
    Thanks for any suggestions you might have!
    -L

    I have never unit tested just a controller. I have always done end to end testing with the systems I have worked on, so I am coming from a different viewpoint than yourself. I would tend to think that just running a full unit test all the way to your database would be the best test. You will be more likely to find the issues with your program if you are testing with an environment as similar to production as possible.
    After running a test you can either purge the data you entered from your test by having sql scripts run against the db, or if this is difficult just reload the database from its original backup. You would probably be able to purge the data quicker than reloading the database from my experience, but it depends on your complexity of the sql scripts.
    best of luck.

  • How can I develope a native mobile cross platform app that can access an SQL server.

    I would do it in html5 as a web app but I also need to access an SQL server and upload data(text) pictures from the camera and
    geolocation data etc.
    I've developed many web based applications, but this is this first attempt to integrate native device functions AND interact with an SQL database.
    So it would have to be able to use form data  then add a photo and geolocation and send everything to the database.
    A site with examples and/or tutorials would be nice.
    To make things worse, I'd also have to be able to work offline and re-sync when networks connections allow.
    If a native app would be too complex, then I was thinking I might be able to develope an app that gathers data then
    passes the info on to a browser to connect with the database.
    But as much as possible I don't want to have to swing back and forth between a browser and an app.
    Pete

    Here is a page a I bookmarked a while back to use jQUERY to access a php scripts sitting on a server that can then interact with the database in the usual way. This means that you will be able to use DW to create the usual HTML CSS and jQuery to make an interactive app.
    I havn't yet tried this but a quick look over the script looks good. Make sure you use php that sanitizes input.
    http://stackoverflow.com/questions/8246380/adding-a-database-to-jquery-mobile-site
    Let me know how you go.

  • Unit test cases run twice on TFS2012 build server

    Hi,
    I have enabled the build definition to run the test cases using Visual Studio Test Runner in TFS2012 build server. Once it runs I can see that each test case ran twice. Suppose if I have 5 test cases, 5 will be passed and same 5 will be failed. Why it ran
    for second time and failed? Please advice.
    Below is my activity log. If you notice at the end of the below log, same test cases ran twice. First all are passed and then failed.
    If Not DisableTests00:00:21
    Inputs
    Condition: True
    Run Tests00:00:21
    If Not TestSpecs Is Nothing00:00:21
    Inputs
    Condition: True
    For Each TestSpec in TestSpecs00:00:21
    Inputs
    Values: Run tests in test sources matching **\*test*.dll, Target platform: 'X86'
    Try Run Tests00:00:21
    Handle Test Run Exception00:00:00
    If testException is NOT TestFailureException00:00:00
    Inputs
    Condition: False
    Set TestStatus to Failed00:00:00
    Inputs
    LabelName:
    SourceGetVersion:
    KeepForever: False
    DropLocation:
    Status: None
    LogLocation:
    BuildNumber:
    Quality:
    TestStatus: Failed
    CompilationStatus: Unknown
    If spec.FailBuildOnFailure00:00:00
    Inputs
    Condition: False
    Get Requests Approved for Check In00:00:00
    Outputs
    Result: IQueuedBuild[] Array
    Mark Requests for Retry00:00:00
    Inputs
    Behavior: DoNotBatch
    Requests: IQueuedBuild[] Array
    Force: False
    If spec Is AgileTestPlatformSpec00:00:00
    Inputs
    Condition: True
    Run Visual Studio Test Runner for Test Sources00:00:00
    Assign spec to agileTestPlatformAssembly00:00:00
    Inputs
    Value: Run tests in test sources matching **\*test*.dll, Target platform: 'X86'
    Outputs
    To: Run tests in test sources matching **\*test*.dll, Target platform: 'X86'
    Find Visual Studio Test Platform Test Assemblies00:00:00
    Inputs
    MatchPattern: d:\Builds\8\Test\OfficeLink\src\**\*test*.dll
    Outputs
    Result: System.Linq.OrderedEnumerable`2[System.String,System.String]
    If Visual Studio Test Platform Test Assemblies Found00:00:00
    Inputs
    Condition: True
    If agileTestPlatformAssembly.HasRunSettingsFile00:00:00
    Inputs
    Condition: False
    Run Visual Studio Test Runner for Test Sources00:00:00
    Inputs
    ResultsDirectory:
    TreatTestAdapterErrorsAsWarnings: False
    UpdateFrequencyTimeout: 30
    RunSettings:
    Platform: Any CPU
    TestSources: System.Linq.OrderedEnumerable`2[System.String,System.String]
    KeepAlive: False
    UpdateFrequency: 5000
    TestCaseFilter:
    PublishResults: True
    RunName:
    Flavor: Debug
    ExecutionTimeout: 0
    ExecutionPlatform: X86
    DisableAutoFakes: False
    Passed TestMethod1
    Passed TestMethod1
    Passed RegisteredTypeShouldBeRegistered
    Passed RegisteredTypeLifeTimeManagerIsTryShouldBeRegistered
    Passed RegisteredTypeWhithoutNameShouldBeRegistered
    Passed RegisteredInstanseShouldBeRegistered
    Passed RegisteredInstanseWithoutNameShouldBeRegistered
    Passed RegisteredInstanseShouldBeResolved
    Passed AllRegisteredInstanseShouldBeAllResolved
    Failed RegisteredTypeShouldBeRegistered
    Failed RegisteredTypeLifeTimeManagerIsTryShouldBeRegistered
    Failed RegisteredTypeWhithoutNameShouldBeRegistered
    Failed RegisteredInstanseShouldBeRegistered
    Failed RegisteredInstanseWithoutNameShouldBeRegistered
    Failed RegisteredInstanseShouldBeResolved
    Failed AllRegisteredInstanseShouldBeAllResolved
    Test Run Completed. 16 tests executed.
    Test Run Failed.

    Hi DHegde,  
    Thanks for your post.
    You’re using TFS 2012 Update 4 and VS 2012 Update 2?
    Please manually build your project and run test project using VS 2012 on your client, then check the test case result in Test Result window, the test cases be executed as expect? 
    If everything works fine when manually run your test project using VS 2012, which build process template selected in your build definition? If it’s not TFS 2012 default build process template, please try use default build process template in your current
    build definition, then queue build definition and view test result in TFS Build log again.
    If you’re using default TFS 2012 build process template in your current build definition, but still received the same issue, how did you configure the test in your build definition, please share the detailed configure screenshot(s) here.
    We are trying to better understand customer views on social support experience, so your participation in this interview project would be greatly appreciated if you have time. Thanks for helping make community forums a great place.
    Click
    HERE to participate the survey.

  • Best Practices for unit testing an Xlet?

    I'd like to be able to unit test and Xlet using JUnit, but have been stumped by how this can be done correctly.
    The first inclination is that I should mock the XletContext, but that seems to be getting out of control quickly.
    Does anyone have any tips on this they can point me to?
    Thanks.

    Hi Jean-Paul Smit,
    Here is a goog article about how to unit test SQL Server 2008 database using Visual Studio 2010, please see:
    http://blogs.msdn.com/b/atverma/archive/2010/07/28/how-to-unit-test-sql-server-2008-database-using-visual-studio-2010.aspx
    Thanks,
    Eileen
    TechNet Subscriber Support
    If you are
    TechNet Subscription user and have any feedback on our support quality, please send your feedback
    here.

  • Unit Testing with Microsoft Sharepoint Emulators and Fakes with Visual Studio 2013

    Hi All,
    I have created Test Project and now creating Test cases for Sharepoint. I found a link on MSDN which suggests using Fakes framework but it supports VS2012 and I am using Visual Studio 2013.
    So how can I use it with VS2013 or is there any other way with which I can implement the Test cases with VS2013.
    Please suggest.
    Thanks in advance.
    Himanshu Nigam

    Hi HimanshuNigam,
    According to your descrition, my understanding is that you want to use Fakes framework to create test case for SharePoint project in Visual Studio 2013.
    If you want to test using Fakes Framework, you can use the codeplex extension to achieve it. It supports Visual Studio 2013.
    Here is a detailed article for your reference:
    Better Unit Testing with Microsoft Fakes
    About how to include the Nuget package, you can use the package with the link below:
    NuGet Package Manager for Visual Studio 2013
    Installing NuGet
    If you still have question about this issue, I suggest you can create a post in Visual Studio, more experts will help you and you can get more detailed information from there:
    https://social.msdn.microsoft.com/Forums/vstudio/en-US/home?category=visualstudio%2Cvsarch%2Cvsdbg%2Cvstest%2Cvstfs%2Cvsdata%2Cvsappdev%2Cvisualbasic%2Cvisualcsharp%2Cvisualc
    Best Regards
    Zhengyu Guo
    TechNet Community Support

  • Hidden, Dynamic parameters in Visual Studio web test - debugging

    Hello! I am trying to debug a web test in Visual Studio and I'm running into the issue of incorrectly bound dynamic parameters.
    Here is the error I am getting: 
    Request failed: Context parameter '$HIDDEN1.j_id71:_link_hidden_' not found in test context
    I have looked at many blogs and websites that tell you how to fix this bug. (http://blogs.msdn.com/b/edglas/archive/2010/03/24/web-test-authoring-and-debugging-techniques-for-visual-studio-2010.aspx) However the websites haven't helped, my error is slightly
    different from the one they outline. 
    This website in particular tells me how to find the place where the parameter is defined earlier in the code, but in my case this parameter never seems to have been defined. I can't figure out what extraction rule to use (if any). 
    Let me know if you have any suggestions/useful websites I can look at! Also let me know if you need more information. Thanks in advance!
    -Danielle

    Hello Jack,
    Thank you for your detailed response. I had already gone through the first blog you recommended to no avail, but the second one seems to be leading me more in the right direction.
    I have a few more questions for you after looking through this advice.
    I added a validation rule like that second blog (http://blogs.msdn.com/b/slumley/archive/2007/11/19/debugging-errors-in-load-test.aspx) says to, and I am seeing that my form parameter "j_id71:_link_hidden_" doesn't exist. 
    However, I noticed when I right clicked on this parameter name in Response tab of the request that is failing, and selected "Quick Find", it searched for the value that the parameter has been assigned. It found this value in a QueryString parameter
    named "_" in the previous request. 
    After studying my web test, I noticed also that this variable "_" is in many QueryStrings, and it always seems to match these hidden parameters that are not being found in the test context. I believe this query string parameter is the main source
    of my issue.
    The validation rule I added from that blog post is trying to validate the existence of a Form Post Parameter, however the value I want to extract is in a QueryString parameter. Is there a way to extract a value from Query Strings? I only have found ways
    to extract from Form Post Parameters. 
    What do you recommend? Here are some things I've tried that have failed:
    1. I tried renaming the parameter "_" to match the hidden one, that didn't work.
    2. I tried creating a new Form Post parameter that matched the Query String parameter in name and value. Furthermore, I tried to use the Validation FormFieldExists rule from the blog to validate that the Form Post Parameter "_" exists, and that
    failed. Why isn't it seeing the form field I create?
    3. I tried to set the value of j_id71:_link_hidden_ Form Post Parameter in the failing request to {{_}}, so that it finds the value from the correct variable. This didn't work either.
    Any other options you can suggest? I know this is a lot of information, I figured the deeper context may help you to help me debug this issue. Thanks again!!
    -Danielle 
    [UPDATE] -- I tried to validate existence of a form post parameter that is obviously there and was there from the beginning, that also failed. Maybe a bug in the validation rule code...
    [UPDATE 2] -- I have started to solve my issues by changing the value of these Form Post Parameters to certain QueryStringParam values. By looking in the Properties panel when I've selected a form post parameter that wasn't working, I can change the value
    to one of these values using the dropdown. The dropdown menu has about 7 different options named "QueryStringParam11810.ct", for ex. Each param has a different number and ending associated with it, and by trial and error I've been finding matching
    QueryStringParams to go with each hidden parameter. So far this is working... Not sure how to choose which value matches in a more specific way besides trial and error.
    So I guess this is resolved, although if you have an explanation for all of this that would be very appreciated!

  • Recommendations for cross-platform software development?

    I'm considering developing a new shareware product and I'd like for it to be able to run on both OS X and Windows, and Linux too if that ends up not requiring too much more effort.
    I've been a Java programmer for many years now, making Java in some ways a very obvious solution. But I've never been all that pleased with the look and the performance of Java Swing GUIs. Swing isn't all that bad for some uses, but it can be clumsy in many ways (like Mac open/save file dialogs), doesn't always get the look-and-feel of a particular platform quite right, misses many platform-specific features one might want to access -- in short, Java-based apps seldom manage to have the fit-and-finish of off-the-shelf, shrink-wrapped commercial software.
    I could stick with Java and try out the SWT toolkit (it's better at getting native look-and-feel, and native performance) for building a GUI, but from what I've read SWT has its limitation as well, and besides, because of the nature of the product I have in mind -- something dealing with audio and music -- a lot of the open source code I'd like to incorporate into my product is written in C or C++, and I might need to do some low-level driver work too. Trying to bridge all of that into a Java app with lots of JNI work doesn't seem like the best route to take.
    I could learn C# and use Mono to make my stuff work for OS X, but, although I've never worked with C# and whatever standard GUI libraries one would typically use with C#, I suspect that it will work best for Windows and be kind of iffy for Mac use.
    All this said, I'm fairly certain that, after many years away from it, the best thing for me to do is get back into using C++ for this project.
    From my cursory preliminary research, the best looking C++ cross-platform development library I've seen so far is Qt from Trolltech. It's also very expensive from the viewpoint of a start-up operation -- a few thousand dollars, even with a 65% discount given to small start-ups.
    There's the totally free, open-source wxWidgets, but it sounds like wxWidgets' Mac support isn't the greatest, and it sounds like it's noticeably buggier than Qt. Maybe it would still be good enough for my uses, but I don't know that yet -- which is one reason I'm writing this post, hoping others have had some experience with it.
    Anyone know any other C++ cross platform toolkits for general GUI application development? One article I read said:
    If your needs are a free cross-platform environment, then wxWidgets is your best and only solution. However, for under $100, you can find a cross-platform environment with a much better Macintosh user experience than wxWidgets.
    I can't seem to find any of these better-than-wxWidgets, under-$100 solutions, however. Suggestions, anyone?
    Quad G5 + 15" MBP   Mac OS X (10.4.6)  

    Hi--
    I think, from looking at that page he must be talking about the CPLAT framwork. It's kind of hard to figure out from that page, but he's got another page with information on CPLAT here.
    It looks like it's $50 for the license. You can also look at the official CPLAT site for more info. I've never used it, though, so I can't really say whether it's any good or not. But there is a trial version available...
    charlie

  • What's the proper way to do this cross-platform app?

    hey y'all
    I'm making a multimedia app  with LV2010 and I'm trying to make it cross platform (win and mac os x).
    The functionality I need requires the use of  system libraries, I have access to somewhat similar functions in both OS's
    (winmm.dll in windows, coreMidi.framework in mac OS X)
    Except for the  actual system calls  the VIs will be identical.
    I can make the VI work in either platform by using the  call library function node,  but  i don't want to  have two copies of everything.
    I've tried  doing  a  wrapper sub  VI that puts  the  call library function node  in a case statement  and detect the OS but this VI will not be  executable because of dependencies (won't find the  library referenced in one of the cases. depending on the OS executed)
    Is there any way  I can specify  which subVI's to load  on a project or something to  get this whole dependency issue sorted?
    Solved!
    Go to Solution.

    You want to wrap the platform-specific code in a Conditional Disable structure, with one condition for each supported platform.  The OS symbol is already defined; see the help for Creating Conditional Disable Structures, which lists the pre-defined symbols and values.

  • Webinar (Aug 11): How to create Cross-Platform Automated GUI Tests for Java Apps

    Join Squish expert, Amanda Burma, and learn how to create cross-platform automated GUI tests for your Java applications!
    Register here (multiple time slots)
    August 11th 2014
    Duration: 30 minutes plus Q & A
    This webinar will cover:
    General Squish for Java overview
    Automating BDD Scenarios
    Executing Cross-Platform Automated GUI Tests
    Interacting with Java application objects, properties & API
    See you there!
    Unable to attend? Register and we'll send links to future events and access to our webinar archive following the event.
    Webinar schedule
    Learn more about Squish
    Evaluate froglogic squish

    <property name="messaging.client.jar.path" value="Location in your local drive" />
    <property name="messaging.client.jar.name" value="nameOfYourFile.jar" />

  • Flash Builder 4.7 Unit Tests for Mobile Devices

    I am trying to setup a simple unit test using Flash Builder 4.7. I created a Test Suite and a Test Case class.
    Using the menu, I select Run > Run. I see options to select which Test Case / Methods I'd like to run.
    But the strange thing is, next I have to select a launch configuration. I'd assume it would just launch in some console window and give me test results. I did what I could and have it setup to run on the AIR simulator for a XOOM device. Nothing happens I just see a blank screen with a blank bar at the top.
    I've also tried right clicking the project and choosing execute flexunit tests. My FlexUnit Resulst still says 0/0.
    What is the launch configuration I need to use to execute the test cases and get the results?

    I solved it! After looking at example code on google, I realized that the s:ViewNavigatorApplication was missing two key options:
    firstView="flexunit.flexui.FlexUnitTestRunnerUIMobile"
    firstViewData="{getFirstViewData()}"
    Thankfully I was able to derive this. It makes 0 since to me why this wouldn't be added to the automatically generated mxml file but the functions it utalizes does. This to me seems like a defect. If you'd like people to use automatically generated files, make sure they are functional.
    Thanks

  • Is Java the only real way of doing cross-platform GUI

    I'm planning a project that will involve some GUI-centric work (its essentially a spreadsheet with some functional programming features). I want to have it be cross platform because I would like to build it as a tool for first time programmers. I haven't really done much GUI programming before but from what experience I have, it seems that the JVM with Swing is the the least painful for the user. I've had some experience with Python, but can't find something that works equally well on all platforms without the user installing a lot of stuff. I've also considered going the JavaScript+browser route, but that's not something I'd like to tackle. Any other ideas?

    Trent wrote:
    Just about anything can be statically linked.  You can even statically link Python and Perl scripts with the interpreter and all the needed libraries, although that would of course hog lots and lots of space that probably isn't necessary.  Compiled languages like C and C++ are the best options for static linking imo.
    I find it contradictory that you want "to give users a single file [that] they can just click and run without having to install a bunch of frameworks first", and immediately turn to Java.  Java is not universal, and it's a significantly large framework to have to install just to run one program.  I can write a GUI program with Python and Tkinter and tell Windows users to install Python first; it's the same thing as compiling a .jar and telling Windows users to install Java first.  If you really just want a click-and-run executable, then your only option (for Windows) is a statically linked .exe.
    Difference being that most probably the user already has java, since more and more webpages require it to run.

  • Does xcode5 CI supports other platforms? like Andriod, cross platform(sencha, appcelarator)

    Hi,
    1) Just i want to know xCode5 CI(Continious Integration) supports other platform like Andriod , Crossplatform(sencha, appcelarator)
        Example: i have andriod project, Can i run the Continious Integration set up with Xcode5?
    2) Does it support large no of build tools? like, Different development teams use a different set of tools, test frameworks, and so on
        Example: codecoverage tool,
    3) What are the advantages of using Xcode5 CI over Jenkins CI
    Regards,
    Venkatesh

    Dear Venkatesh,
    You have asked your question on a Server forum whereas Xcode is really a developer tool and the conversation is more suited to the developer forum.  However,
    1) Xcode is only useful for developing for Mac OS and iOS.  Technically it can compile code for other platforms like Android using the standard compiiler but it's not really designed for that and I wouldn't recommend starting a project that way.
    2) Xcode is an IDE but it comes with (free) developer tools like the ones you mentioned, for unit testing, diagnosis of processor and memory usage, corruption testing, instrument testing and other purposes.  Also included is a simulator for iOS devices so you can debug your iOS code for all the different devices without needing examples of them all.
    3) For Mac OS and iOS, the combination of Xcode and Xcode Server has more and better tools than Jenkins, but, as answered in (1) it's useless for other platforms.

  • Unit testing or integration testing?

    Here it says that this example is more of an explanation of framework than best practice because of the conflation of integration/unit testing concepts: http://docs.flexunit.org/index.php?title=Using_Asynchronous_Startup
    My question is, how do you define a line here between unit testing and integration testing? What kind of UI components asynchronous tests are then considered to be unit tests and not yet integration tests? Many books recommend having 100% code coverage with unit tests. How would this example look like, to be considered only a unit test and not integration test?

    The question about which levels of testing to do are really dependent upon you. Let me explain a bit more.
    If you are building an application, and only an application, then functional testing probably covers most of your needs. In an application environment, you have a relatively narrow pathway of options and interactions that a user might follow. So long as you test those options well, your app will probably work as well.
    Where I see the real value of unit testing is testing for the unknowns. If you have a class that will be used in multiple applications in multiple ways, or perhaps it is intended to be a reusable class where others in your organization will use it in the future, then you want lower level testing. You want to have the best coverage you can because you want to know that, regardless of the way someone chooses to use the code, it will likely function or at least fail gracefully.
    The same is true for integration. If I am creating a Flex component, then I can unit test part of it, but not all. So, I want to use integration tests to ensure that it works in as many circumstance as possible. Further, Flex components have a life cycle, so testing them when they are not on the display list isn't really a valid test..
    So, to summarize, I see unit and integration w/Flex about testing possibilities and trying to achieve realistic code coverage for known and possible use cases. I see functional testing as testing those components in very specific use cases.
    Which do you need? Depends. Are you developing reusable components across (n) projects, likely the first 2 and maybe the last. Are you an application developer testing that an application works in documented business cases? Just functional will likely work.
    Hope that helps,
    Mike

Maybe you are looking for