Hi Cliff...

I've just added you as admin

regards

:::poly:::

Edit : sorry Cliff... i had a problem yesterday.. you are now admin

Facebook >.<

People still use that?

Well... at least me... What kind of social network do you use ? (if you use one)

:::Poly:::

Hi,

I've noticed that there is no PureMVC pages on Facebook... so i've created one...

http://www.facebook.com/pages/PureMVC/128081860951

Cliff and other contributors can add me (Francois Polykrom Dejardin) and i'll added you to administrators...

Regards

:::Poly:::

Hi,

so, if it's just to be sure your comboBox content is loaded, you can use StartupManager and, when the content is loaded, send a stateMachine.ACTION notification to swap to the comboBox view and use a bindable var wich contain the combo values to populate the component...

Hi,

If you are in a StateMachine logic for the main application, you could use StateMachine in all other "components" or Modules... Each Core can have its own StateMachine (you have to use the multicore package for that).

The way you use StateMachine seems not to be the right way.. You should have one State Machine for your main app with two views LOGIN and POPULATE COMBO.
And in each view, other StateMachines or Notifications handlers to swap between components view.
 
For your FSM, i think the one you provide would be for the Project loading module... not for the main one.. and don't forget to define the target state in your FSM :

ApplicationFacade.STARTUP_FAILED,ApplicationFacade.PROJECTS_LOADED and ApplicationFacade.PROJECTS_LOADING_FAILING

regards

:::Poly:::

Hi,

difficult for me to say what is the more difficult things to understand in PureMVC now that i understand the main framework "rules"...
But,I think that for all stuffs we focus on, we have to work hard to understand all the concepts and best practices.
The first thing that comes in mind is (as for other frameworks) to understand that even if we have first to make a lot of codes for little tasks, it's to make the applications more robust and extensible...
After that step, i think that (when understanding main design patterns and other frameworks like Cairngorm) PureMVC give us in two packages (one core and multicore), all the architecture and tools we need to make our life easier !
When i first began to work with pureMVC... while i was progressing in my development, i found all the answers i need to continue (utilities, forums examples..) .
The support forum is great and there are a lot of basics and more complexes examples to learn.
Authors are always available to answer our requests... It's very motivated..

So.. What makes PureMVC a great framework is that when we put our hands in it we found things naturals, everything seems more clear and more "real life" oriented in terms of communication and architecture.

In conclusion, i think that Jason is right in the "Hardest" concepts to learn , and i think all the utilities are essential to realize the real benefit of PureMvc.

my €0.0001

:::Poly:::

Just to remember that Flex Framework can also be implemented in RSL, so it decrease also the size of the application itself and modules...

This link is very usefull : http://livedocs.adobe.com/flex/3/html/help.html?content=projects_7.html

my two cents

:::Poly:::

In Flex Builder, when building your module, in the properties page for included libraries, select 'external' instead of the default 'merged into code. Why? Think about it. If you have two modules and the shell, and they will all use the same VO, then if you compile them with the 'merged into code' option, you'll be loading the same class three times by the time the third module is loaded. Generally not desirable.

And it is not always true that the shell should be the smallest and the modules larger, in fact it may often be the other way around, since the shell often includes all the shared stuff *including* Flex, PureMVC and Pipes.

-=Cliff>

Yess Cliff, i was just answered when you post

I've understand the process now.. and it's true that my modules are not smaller than the shell... Have to finalize some piece of work to see...

Anyway, thx a lot to Tek and Cliff

chuss

:::Poly:::

Ok, i finally found the solution myself...

I've created a common library with all my common classes, assets and so on...(also all PureMvc Framework and Utilities Classes)
I've added it in my main application Library as RSL, so i now have a new .swf library that is shared between shell and all my modules.
The weight of each module is now divided at least by 2 !

Wooow... so cool !

thanks

:::Poly:::

I consider that it may be possible to run any module alone, without the core at anytime as much as possible to guarantee a good implementation.

Thanks Tek.. i'm on the right way so...

but, when you say that each module may run alone, in my case, it's not possible cause i get content and relation between shell and modules with pipe... so when i launch my module .swf, it doesn't display anything..

And following Cliff advice... Have i to put all application components in an external library ? I don't understand what Cliff said about not Merging library in Module (Have i to create Application for each modules with same external lib not merged into code and Shell application with the module libraries merge into code ??)

regards

:::Poly:::

I've just read that in other posts :

Just create a new Flex Library project to put the shared classes in.

At a minimum, you'd have three projects:

1) Your main app project,
2) Your module project
3) Your library project.

You'll make both your app project and your module project include the library project in their build paths.

-=Cliff>

The Modularity demo has a lot of ground to cover, and the primary focus of it is the communications by interface between the shell and the modules.

In a real project you can separate out your modules into separate projects, create a shared 'API' project that has just what needs to be shared between the modules and the shell. Then the client builds in the API library and loads the modules. The modules should be compiled with the API being 'external', not 'merged into code'. The main app merges the API into its code and provides it to loaded modules, so they don't have to be compiled with it.

-=Cliff>

Have i to put my common Vo, shared components and so on in a library ? Just have to test it... because for the moment my Modules are larger than my main Shell (It should be the opposite)

Also, when Cliff say "The modules should be compiled with the API being 'external'", have i to create a Flex Application for each module ?

thanks

:::Poly:::

Hi,

As i said in a previous post, I've successfully integrate Pipes in my PureMvc Multicore application.

In my Startup shell,i get Config and Content of my application, I use StartupManager to load 2 XML files (Config.xml,Content.xml), these xml are parsed and put in 2 values objects. At this point, no problem.
The two vo are accessible by my shell main container, I then load different cores modules that have their own facade and so on... but these modules need the content too...and i've also need to update shell vo from my modules.
I've read some posts here and best practice seems to not considere the Shell to be a "grand station central".. and it makes sense..
Actually when i load a module and check that this module is initialized, i send a notification from the shell to module pipe with the content vo...
Is it the best way to get and post global vo between Shell and Module or have i to create a centralize module to keep all (It would be the same issue that the shell solution).

Thanks for your advices !

:::Poly:::

Hi Cliff,

Don't worry about that, i can imagine you have a lot of work to give us such fantastics stuffs !

I've successfully implemented StartupManager,AsynchCommand,StateMachine and Pipes in my MultiCore PureMvc application. And it rocks !!!

Everything is clear now and will help me to win a lot of time and made my applications more robust and extensible.

Keep the good job

regards

:::Poly:::

Ok... sorry for that post,

I've finally found my answer in a previous post ... One stateMachine per Core...

http://forums.puremvc.org/index.php?topic=1318.0

and the Pipes philosophy is what i need (the PipeWork demo is a great source of inspiration)

MMmhh... watching http://forums.puremvc.org/index.php?topic=859.0 and it seems sub states isn't possible without subClassing FMSInjector.
Is there any example of that way ?
Have i to create facade for each of my components ?

thanks

:::Poly:::