|
|
Show Posts
|
Pages: [1] 2
|
5
|
Announcements and General Discussion / Architecture / Modes
|
on: September 17, 2009, 07:01:24
|
So I have an app that will have, in the near term two modes, edit and view mode, but could have 3 or 4 more modes in the future. Really not sure if modes is the correct term, think Photoshop or Illustrator. The user can click a tool and do certain things in my app so with a hand tool a click might mean to open an object; view mode. In edit mode, that same click will could be something like add an object. I am wondering how this fits into Puremvc? I would rather not put messy conditional logic in giant mediators, so would it be better to create dual mediators? Or is there another way? My app is huge (think upwards of 1000 UIComponents) and takes some time to start up to render everything, so I don't really want to use the multicore and open a second version of my app for editing. What are my options? What about making the mouse pointer/tool be "ALL MIGHTY"? Think photoshop. What I mean by almighty is, when you select a tool, the toolmediator recognizes the switch and updates the toolproxy to change the tool value. Then the tool would send out different toolEvents based on its state (aka what kind of tool it is) the toolMediator would thus send out notifications based on the type of toolEvent it receives? See any downsides to this? TIA, Jason
|
|
|
6
|
Announcements and General Discussion / Architecture / Memory Usage and Multi-core
|
on: June 29, 2009, 08:49:15
|
So my app is up in beta, but now I get to do a major rewrite....maybe.
So I have an app, built on Single version of PureMVC, which can be extremely processor intensive. I let the user switch between two views, possibly five views in the future. When the user switches between the two views I remove all the proxies, mediators, etc... and build the new view from scratch. So I was thinking about moving to multicore to allow for a document based application so the user can have tabs open (like Firefox) and switch between views.
The question I have is, since a single view right now can eat up memory and cpu, would using modules and multicore in a document based app be ok or would it kill the CPU and eat more memory? I would think if I put the views in a tab navigator it the views that aren't being shown wouldn't eat up as much memory if they aren't being rendered and not animation or data processing is going on in those views, right?
Are modules the best way to go here? Maybe it would be better to create new view objects in a tab navigator, rather than modules with view objects in each tab navigator? Honestly, I hate dealing with modules, but if it will give the best performance then so be it.
Maybe sticking with what I have would be the best way, given the processor and mem usage of the app?
|
|
|
8
|
Announcements and General Discussion / General Discussion / FLEX GUMBO ISSUES
|
on: June 16, 2009, 09:37:52
|
Wondering if anyone is using PureMVC with Flex Gumbo yet? I have been using the beta or alpha for about 3 months now and my trial expired today. So I installed the official release BETA and changed all my namespaces over to the new ones, compiled the app, and half my app doesn't work.
For instance, I have a stage mediator that on a certain action I call a function <Application> from the stageMediator that creates a popup. But now on the new BETA that function is called and it never reaches the function in <Application>.
Any suggestions?
|
|
|
9
|
Announcements and General Discussion / Architecture / Backed into a corner with Single Core
|
on: May 22, 2009, 01:43:31
|
So I have this fairly large Flex/AS3 application built on Single Core PUREMVC. I was 95% done with the project (think google maps for it networks) and now they want a new view, which does half what the old did and half a bunch of new stuff. So what I tried to do is keep Single Core version and when they switch views send a TEARDOWN notification so all mediators and proxies get removed. Then I create the new view, but half my functions don't work and I have a feeling the new view is getting hold of the old mediators somehow? I can't seem to track down why things aren't working, for instance in my displayMediator I have a reference to something on the mainstage and when I try to reference the mainstage in the new displayMediator it comes back null. So knowing its a google maps type app where I have a couple of zoom / pan controls which always need to be on top of the stage what do I do next? Would switching to multicore be the solution please say no as I can't imagine how many hours its going to take to transition everything over? Help me please , Jason
|
|
|
10
|
Announcements and General Discussion / General Discussion / ServiceLocator RemoteObjects?
|
on: May 21, 2009, 07:37:42
|
So how would you implement a Cairngorm style servicelocator in Puremvc?
I am thinking you would have some sort of "Main" proxy, for instance, I have a stage mediator that listens to the stage for certain things so maybe a StageProxy would be the best location for this servicelocator or would I create some sort of ServicesUtil.as that the StageProxy would then use to its liking?
|
|
|
11
|
Announcements and General Discussion / Architecture / State Machine???
|
on: April 29, 2009, 12:56:41
|
So I have a Flex application that will have up to 1100 items on screen at all times (zoomable) and for right now they are all static, meaning once on screen they won't change. But, in the release we need to have each of the object be dynamic, meaning that depending on the VO they are passed they need to update their color, size, shape, and position on the stage. Some may even disappear be removed all together and some new ones added depending on the data thats pushed in. The user will also be able to drag the items to different positions on the screen.
Also in the next release, when the user leaves the app, I need to save the state of each of the 1100 items (their position on the stage, color, shape, size, etc...).
So I am thinking local shared object, but do I or can implement the PureMVC StateMachine? Or is their a better way to do this?
Also I am using the Single Version of PureMVC.
|
|
|
12
|
Announcements and General Discussion / General Discussion / Mediator already exists
|
on: March 30, 2009, 11:54:59
|
If a mediator and proxy already exist for viewComponent will a call to facade.registerMediator create another one and destroy the old one or will it just pull back the old one?
Is it better to do something like
if( facade.retrieveMediator( "mymediatorname" ) ){ // do nothing skip creating a new one. }
|
|
|
13
|
Announcements and General Discussion / General Discussion / How to handle ONLY ONE callout
|
on: March 25, 2009, 04:29:01
|
So I have a Puremvc (Single) / Flex app and in this app I have 'Callouts', which are like Google Maps callouts when you click on a marker. Just like Google Maps I only want to allow one Callout open ever. Just like Google Maps I if my app is dragged I need the Callout to follow so right now when my callout is created I addEventListener for enterframe event and then send my Callout its new coord, which are my markers new location.
All this works fine, but the issue is I cannot seem to find a clean way to remove the callout and its eventlistener when a new node is clicked. I have a very dirty hack that more than likely breaks PureMVC best practice right now and of course it isn't stable, thus I am posting.
And no I cannot use the Flex popup, this is for several reasons that would take to long to explain. Also to make matters worse my callout is on the main stage because it needs to be on top always and my markers are down about 4 levels in the display list.
All that being said does anyone have any suggestions on the best way to handle this? The issue again seems to be in removing a callout and then immediately creating a new one. I have traced the issue back to UIComponents.callLaterDispatcher. When I click a new marker I send a notification that ends up in a callout being closed, but then immediately creating a new one and UIComponent ends up storing all this in a queue that I guess is getting backup in workload. Or it could be something else. Anyway i figure via PureMVC their is a better way to do this. Would I be correct? Maybe something like not removing the eventlistener and callout but hiding it and then just assigning it to a new marker and then turning it back on?
Though when I do want to close it manually, not by clicking another marker how would it get removed?
Hope I didn't lose you.
TIA, J
|
|
|
14
|
Announcements and General Discussion / General Discussion / Proxy referencing other proxies
|
on: March 23, 2009, 03:23:40
|
Is it bad form to have proxies referencing other proxies, so as to have access to the retrieved proxies data? mybigproxy{ facade.retrievePorxy(someproxyname); }
Or would it be better to use a command to create a proxy and send in the data from another proxy then.
execute { facade.retrieProxy (myproxy); facade.registerProxy( new Proxy( myproxy.getData() ) ); }
|
|
|
|
|