[FlexJS] Road Show

2014-03-18 Thread Alex Harui
Hi Folks, Apologies if you've already seen this on the dev@ listŠ You may have seen mention of FlexJS, a version of Flex that can cross-compile MXML and AS to HTML/JS/CSS so the app can run without Flash. Most discussion about FlexJS happen on the dev@ list so you are encouraged to subscribe to

Re: event whenitem render is complete

2014-03-18 Thread Javier Guerrero García
A workarround here: in your itemrenderer, call the sizing function if data = dataprovider's lastitem :) P.S. Are you sure you can't do the sizing math in advance? If you adjust the sizing at the end, it'll surely cause a re-rendering On Tue, Mar 18, 2014 at 9:34 PM, Alex Harui wrote: > Most co

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Justin Mclean
Hi, > Maurice, anyone can easily enable telemetry with Renaun's Telemetry > enabler[1] that you can setup and run in in CommandLine to quick enable > telemetry without requiring an updated version of the compiler. The advanced-telemetry flag works in Apache Flex 4.9 and above. Thanks, Justin

Re: event whenitem render is complete

2014-03-18 Thread Alex Harui
Most components dispatch updateComplete, but this looks like a custom ViewStack so I don't know how it actually works. On 3/18/14 11:17 AM, "Scott Matheson" wrote: >is there an event when a itemrenderer is complete i.e. all items has even >rendered from the data provider, i want to size the view

event whenitem render is complete

2014-03-18 Thread Scott Matheson
is there an event when a itemrenderer is complete i.e. all items has even rendered from the data provider, i want to size the view stack when all items have been added minHeight does not seem to work, i end up with 10 lines when i only need 5 Disclaime

Re: Migrating from BlazeDS to GraniteDS

2014-03-18 Thread dude
I would like to convert our current project to GraniteDS & Tide, if it wasn't for the incompatibility with Parsley meta tags ([Inject], etc). :( Am 18.03.2014 17:19, schrieb Franck Wolff: > Hi, > > We have republished and updated a quick migration guide from BlazeDS to > GraniteDS on our blog her

RE: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Maurice Amsellem
Yes, I pressed the button and the result was the yAxis updated with .1 values (around ten ticks). And it was instant. Maurice -Message d'origine- De : DarrenEvans [mailto:darren.ev...@allocatesoftware.com] Envoyé : mardi 18 mars 2014 18:20 À : users@flex.apache.org Objet : RE: Patchin

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread DarrenEvans
As I mentioned, I have inherited this code and the interval was already being set. I do not see the need for an interval of 0.1. I think it will be ok to remove it. The 0 to 100 min/max is purely for example. In reality the data will range from 0 to 8 (maybe up to 10) so an interval of 0.1 is appr

RE: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread DarrenEvans
You did press the button? It's pressing the button that takes 5 seconds to initialise the axis. In the main application with the 4 charts it was taking 30 seconds when the interval was 0.1. In the test application pressing the button takes 5 seconds for the axis to update. Removing the setting of

RE: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Maurice Amsellem
>I haven't found a way to get call counts in Scout. Yes, I agree it's missing. I don't understand why they didn't put that feature. It's critical for any serious performance investigation. Maurice -Message d'origine- De : Alex Harui [mailto:aha...@adobe.com] Envoyé : mardi 18 mars 2

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread DarrenEvans
Very useful post. The "hassle" I spoke of in my previous post was exactly this problem, advanced telemetry. We got round it eventually but that was the problem. Going to install Scout and see where the other 5 seconds are going. João Fernandes wrote > Maurice, anyone can easily enable telemet

RE: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Maurice Amsellem
I have tested the code your provided with SDK 4.6 and linearAxis.interval = 0.1; And it's not 30s or 5s, it's instant. PC: Window7, core I7 2.1Ghz, flash player 11. Do you confirm the same example on your side takes 30s, or is it the main app with 4 charts ? Maurice -Message d'origine

RE: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Maurice Amsellem
>Maurice, anyone can easily enable telemetry with Renaun's Telemetry enabler[1] Didn't know that. Thanks for the information and the link Maurice -Message d'origine- De : João Fernandes [mailto:joaopedromartinsfernan...@gmail.com] Envoyé : mardi 18 mars 2014 17:27 À : users@flex.apache

Re: Migrating from BlazeDS to GraniteDS

2014-03-18 Thread Alex Harui
On 3/18/14 9:19 AM, "Franck Wolff" wrote: > >AFAIK, BlazeDS is dead for about 3 years. BlazeDS code will be in Apache shortly, so it isn't dead. Maybe dormantŠ -Alex

Re: Migrating from BlazeDS to GraniteDS

2014-03-18 Thread Gary Yang
For my experience, it would be better to keep this in mind when using GraniteDS: there should be a layer where DB entity convert in/back to DTO, so that UI logic won't be locked with DB schema. thanks. On Tue, Mar 18, 2014 at 12:41 PM, Justin Ransom Dallas < justindal...@gmail.com> wrote: > I'

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Alex Harui
How big is this chart? 0 to 100 at 0.1 is 1000 intervals. Do you really need that many? On 3/18/14 9:28 AM, "DarrenEvans" wrote: >It's showing on my post here.Don't know whether it got removed >because it >was in a CDATA block > >It's simply: > > import mx.charts.LinearAx

Re: Migrating from BlazeDS to GraniteDS

2014-03-18 Thread Justin Ransom Dallas
I've always found GraniteDS to be far more involved than the BlazeDS lib. Just my opinion. On Tue, Mar 18, 2014 at 12:19 PM, Franck Wolff wrote: > Hi, > > We have republished and updated a quick migration guide from BlazeDS to > GraniteDS on our blog here: > > http://www.granitedataservices.co

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread DarrenEvans
That is a very useful post! The last time we played with Scout the whole advanced telemetry thing was the "hassle" I was talking about in my previous post. We did get it working in the end but I remember it not being a simple case of installing Scout and hooking it up. Whereas the first time I us

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread DarrenEvans
It's showing on my post here.Don't know whether it got removed because it was in a CDATA block It's simply: import mx.charts.LinearAxis; private var requiredVsActualDataProvider:Array = []; private function button1_clickHandler(event:Mouse

Re: Considering apache flex as Enterprise UI for Desktop, Mobile and ipads industry wide

2014-03-18 Thread Alex Harui
FlexJS is a prototype. We hope to create an "alpha" release soon. We are looking for people to get involved with testing and development. Because Apache is a volunteer organization, roadmaps are not very useful. The future of Flex and FlexJS is in the hands of those who have time to contribute.

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread João Fernandes
Yes Alex, clearly it's what's missing in Scout and I've reported it a long time ago (with many other enhancements) but you have an idea how much it consumes in that particular function. Of course if we had consumption average per call would be even better. Like Darren I was unable to run the profil

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread João Fernandes
Maurice, anyone can easily enable telemetry with Renaun's Telemetry enabler[1] that you can setup and run in in CommandLine to quick enable telemetry without requiring an updated version of the compiler. http://renaun.com/blog/2012/12/enable-advanced-telemetry-on-flex-or-old-swfs-with-swf-scount-

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Alex Harui
I haven't found a way to get call counts in Scout. I still find the FB profiler's call counts a very useful tool in figuring out the root cause of many performance problems. -Alex On 3/18/14 8:59 AM, "João Fernandes" wrote: >A faster way to check the bottlenecks in performance is to use Adobe

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Alex Harui
I think the guts of the script block is missing... On 3/18/14 8:39 AM, "DarrenEvans" wrote: >Here is full code to highlight the problem. This example takes about 5 >seconds when you press "Setup Axis": > > >http://ns.adobe.com/mxml/2009"; > xmlns:s="library://ns.adobe.com/flex/spar

Migrating from BlazeDS to GraniteDS

2014-03-18 Thread Franck Wolff
Hi, We have republished and updated a quick migration guide from BlazeDS to GraniteDS on our blog here: http://www.granitedataservices.com/2014/03/18/migrating-from-blazeds-to-graniteds/ AFAIK, BlazeDS is dead for about 3 years, as well as Spring-Flex. On the other hand, GraniteDS is still very a

RE: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Maurice Amsellem
You are right, But in this case, they are using 4.6 and Scout requires the - advancedTelemetry compiler option. Not sure that the compiler that comes with SDK 4.6 supports -advancedTelemetry. That's why I suggested the "old" profiler... Maurice -Message d'origine- De : João Fernandes

RE: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread DarrenEvans
Here is full code to highlight the problem. This example takes about 5 seconds when you press "Setup Axis": http://ns.adobe.com/mxml/2009"; xmlns:s="library://ns.adobe.com/flex/spark" xmlns:mx="library://ns.adobe.com/flex/mx">

RE: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread DarrenEvans
Doesn't need any data, requiredVsActualDataProvider is a blank array! private var requiredVsActualDataProvider:Array = []; I've ruled out actual data being the problem. Can't running the basic profiler as it just crashes because the application is so bigand not set up Scout in a while (last

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread João Fernandes
A faster way to check the bottlenecks in performance is to use Adobe Scout instead of the Performance profiler On 18 March 2014 15:19, Maurice Amsellem wrote: > SDK 4.9.1 chart optimization was on style initialization, which was done > for every chart , while it was required only once per module

RE: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Maurice Amsellem
SDK 4.9.1 chart optimization was on style initialization, which was done for every chart , while it was required only once per module. So the performance gain is only when you have many charts. You say that your test has one chart, and it takes 8 seconds to initialize the axis. - Can you run th

Re: Considering apache flex as Enterprise UI for Desktop, Mobile and ipads industry wide

2014-03-18 Thread Gary Yang
I would say even for Flex 1.5( Flash player 7 ), it's not possible to migrate to HTML/JS, simplely because the difference between the fundamental platform. Again for me what makes FlexJS different from other JS solution is that FlexJS can migrate to Flex( Flash Player ), so that when HTML/JS not e

RE: Considering apache flex as Enterprise UI for Desktop, Mobile and ipads industry wide

2014-03-18 Thread Katariya, Piyush P
Yup. FlexJs is the only reason I am thinking adopting Apache Flex. But wanted to know the current status and further roadmap if any. Regards, Piyush Katariya Senior Product Engineer United Health Group -Original Message- From: Gary Yang [mailto:flashflex...@gmail.com] Sent: Tuesda

RE: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread DarrenEvans
Hi Maurice, Thanks for helping out too. Moving to anything above 4.6 is impossible without severe development from our end. Our primary limiting factor is we use the IBM ILOG Elixir Gantt Chart component and it is critical to our application. This was not originally owned by IBM and was well sup

RE: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Maurice Amsellem
I just tried building a simple project with a chart and swapping the chart.swc. I used -static-rsls=false in the main app so that RSLS are not used , and set linkage to "Merged in code". The application starts and the chart is displayed! Precautions to take: - test "Merged in code" as framework

Re: Considering apache flex as Enterprise UI for Desktop, Mobile and ipads industry wide

2014-03-18 Thread Gary Yang
Great question. I think FlexJS will have a bright future because it can migrate to Flex(FP); On Tue, Mar 18, 2014 at 8:10 AM, Katariya, Piyush P < piyush.katar...@optum.com> wrote: > Hi, > > Recently came to know about apache flex and significant > improvement made after year

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Alex Harui
Hi Darren, What change in Apache Flex do you think or know will solve your performance issue? It might be easier to just do monkey-patching. I don't use IntelliJ so we might need their help otherwise. To monkey patch, just create the folder structure for whatever changed files you want and put

RE: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread Maurice Amsellem
FYI, this is the ticket originating the chart performance improvement : https://issues.apache.org/jira/browse/FLEX-33216 It was fixed in Apache Flex SDK 4.9.1, which is much closer to SDK 4.6 than the current 4.12. You say you are stuck to Flex SDK 4.6, so maybe moving to 4.9.1. would be easier/

Considering apache flex as Enterprise UI for Desktop, Mobile and ipads industry wide

2014-03-18 Thread Katariya, Piyush P
Hi, Recently came to know about apache flex and significant improvement made after year 2011 with subsequent releses. I had a query about the followings: 1. Future of apache flex ? 2. I have heard there are significant efforts towards generating the

Re: Patching Flex 4.6 SDK chart functionality

2014-03-18 Thread DarrenEvans
Hi Alex, When I say "modules", I'm referring to IntelliJ modules. This is a whopper of an application! There is only 1 resulting SWF file but it comprises of many modules (libraries). We have a product platform so have the following base modules: 1. Controls (lib) 2. Resources (lib) 3. Framework

RE: "StageTextInput can now be included in scrollable content" does not work?

2014-03-18 Thread Maurice Amsellem
I have checked again, the hardware button is handled correctly on Android (ends editing). The stack trace indicates suggests that proxy couldn't be created, which is odd. When you create the JIRA ticket, will you please attach the full stack trace, and indicate if you are using 4.12 GA or lates