yup, that's exactly what i did, except I'm not using flex. I stored all my different type of vo's in arrays and then stored all my arrays of vos in a dictionary in the data of my main proxy. Think I got it figured out, thanks as always cliff.

so, say I have one array of 100 VOs. stored in a proxy.

I was thinking I create a "live data" proxy that will hold search results data, have a command that filters through all of the vo's to the ones i want, and then sends a note to update the live data proxy, which refreshes my mediator. So basically having a proxy that always stores all the vos, and another one that basically stores a list of the ones I want to show up in my mediator.. is this what you'd consider the right way to do a simple search?

hey, yeah, I have 5 kinds of VO's with indexes all being stored in multiple dictionaries right now, a single proxy managing them all. There are between 50 - 200 of each kind. I never thought about having a proxy for each VO, that's interesting. I'm trying to think of how to accomplish that. Basically inside my service proxy that is currently creating the vo's, I just tie each one to a new type of proxy with the same id, and then super the data...?

So basically, I have a large set of VOs being created from a remote service proxy. I also want to be able to update this data every couple minutes or so. I make new calls and get back the data, but my question is what would be the best method to only create VOs if they don't exist already?

I feel like it would be very inefficient to check one property of each VO as it's created. Thanks for any tips.

So that you can store the vo's you're retrieving in the models data. Man, I should really stop asking questions on here. But if you have any recommendations on loading the php that would be swell.

I know this may label me as a total noob, but I've just yet to have to do it this way, most of my projects have been driven by our own simple XML.

My question is why is making a remote connection considered a proxy when I think of it as a command?

But beyond that, I would love any reccomendation of the best way to call a php class that returns XML. I've found a couple things online already such as amfphp, which looks awesome and easy, but is there any point to calling binary when the php is sending it formatted as XML already?

As always, thanks, You guys rule.

Haha, yeah thanks, I totally figured it out and thought it was pretty obvious, but hopefully this will help someone else searching in the future . then I realized I didn't even need to do it that way and could access my main app and stage earlier just by creating those mediators first. Problem solved.

ignore me, it was a stupid mistake.

mainly because I want to handle a flash vars proxy in the configure command, but it has no reference to the main app, because that was only carried to the startup command.

I should also mention I am using fabrication, but I don't think it changes anything.

So, I'm learning this again, after a brief hiatus. I can finally use it for work!

Anyway, I thought I had a good grasp on a lot of puremvc, but have come to what's probably a stupid question.

so i'm using a state machine to initialize, and startup my app. Can you pass a note with a body into commands triggered by the FSM???


<state name={CommonConstants.STARTING}>
      
                     <transition action={CommonConstants.STARTED}
                                 target={CommonConstants.CONFIGURING}/>
                  
                     <transition action={CommonConstants.START_FAILED}
                                 target={CommonConstants.FAILING}/>
                  </state>


so when it runs the configure command, is there any way to populate the note?? I'm kind of confused, thanks!

This stuff was way over my head like a month or two ago. I've probably reread the best practices document completely 3-4 times.

But I just wanted to say that once it clicks, it is amazing. I am now close to finishing my first successful pure as3 puremvc app. It's not that complicated, but it's amazing how many stupid problems I've had in the past that puremvc solved really easily.

State machine, is aweesssoommeee.. and that you can use a state machine anywhere in your app too! Not just the initialization. I know, obvious, but it wasn't for me at first.

I haven't gotten into the loadup utility yet, but I sure will eventually. I'm just happy I have a multicore app working very well..

THANKS A TON!

Just wanted to mention that this thread was very helpful for me. Also coming from a pure as3 background.

When overriding execute in my startup command, how does it receive the note? Is the execute inside the startup command linked to the startup notification that fabrication is sending automatically? It just hides this? I'm not having a problem with it, just trying to understand where it's getting the note from, which contains my document class in it's body.

Other than that, just want to make sure I'm building in a proper fashion. Right now I'm almost done with my shell.

This is how it's set up, Startup command registers basic FSM commands. Injects FSM. and then registers my proxies and mediators. then they're configured. then in assembly, I create my different modules... I imagine down the road, my modules will only have two notifications in common with the shell.. address change and page change.. and the rest will be specific to each module.

should I add my modules to the stage in my application mediator?,.. does that sound right? Cliff suggested wisely I instantiate them only and then have each wait for it's appropriate note to initialize. This is what I'm going for.

Once they're added, each module will be able to add view components/display objects right? through each modules' own document class stage.. which would really be a reference to the shell's stage?

appreciate any help greatly.

sorry I put this in the wrong forum.

excellent! the sea of arrows code is a great starting point. as I learn I have to remember to keep coming back to things i once didn't understand. Time to make my first state machine!

Hi, I'm having a great time learning PureMVC, and it's so awesome. Thanks Cliff!

So my question lies with creating modules. I have my application shell that is a simple xml controlled structure. And I also created a slideshow module. My shell is using a version of a swf address mediator as the main dynamic notification structure. The thing I for see happening, is that if someone goes directly to a deep link, and my slideshow module hasn't been loaded yet, it obviously won't be able to react properly to the page change notification. So does this mean that I should load and create all my separate modules initially before my address mediator? In my startup command or my initializing mediator? I assume that would be the best way to handle it, but also liked the idea of loading the modules when they're needed, which sounds proper. Perhaps I have a command that checks the deep link and loads the appropriate module? I'm sure there are a lot of ways to do it, and I could use some advice.

Much appreciated.

Russ