Thanks for looking. There's going to be an RC3 shortly with some installer
changes and some other "minor" things.
From: Mark Kessler [kesslerconsult...@gmail.com]
Sent: Tuesday, July 1, 2014 4:46 PM
To: Dev@Flex
Subject: Re: [VOTE] Apache Flex SDK 4.13.0 R
Hi,
I'm working on creating a way to publish vanilla Flex SDK projects to
JavaScript on latest gen browsers. This project consists of several sub
projects, and I'm wondering what is the best way forward with regard to
contributing them:
1) two new namespaces and accompanying projects in the main
I think the reason for the overlay is that by this we are tricking the
FlashBuilder to use Falcon instead of the old compiler.
It should be possible to have Falcon separate from the normal compiler. It's
just that FlashBuilder will probably not be albe to use it that way.
Probably IntelliJ would
Hi Chris,
Thanks for the feedback.
I'm talking about FalconJX, not Falcon. The latter is the "new" SWF
compiler, the former is the Flex to JavaScript cross compiler.
I guess a feature branch might work... But I'm afraid that keeping that
branch up to date with the 'develop' branch will add extra
No effort at all ... that's what GIT rebase is for.
You can think of this sort of an automated "Create-Patch, Revert, Update, Apply
Patch" ... if all goes well, it's just this one command, if there are
conflicts, you get the usual conflict editor you would get anyway if you
created conflicts on
Due to 14 beta bug. RC3 due shortly
On 6/27/14 11:18 PM, "Alex Harui" wrote:
>Hi,
>
>This is a Apache Flex 4.13.0 release candidate 2. Please see the
>RELEASE_NOTES and the README.
>
>There is a known bug in Adobe Flash Builder 4.6 and 4.7 that causes a
>issue when creating new applications.
>
Please vote to approve Apache Flex SDK Installer version 3.1 to be
released.
Issues addressed in this release:
- Installer will report more information when installs fail.
- Bugs fixed in ant_on_air required by latest nightly builds of FalconJX
and FlexJS
- Caching logic moved to the install scrip
This is the discuss thread.
This RC helps fix the AIR 14 beta download issue
I also added a -log option to log to a file.
Changes are minor so votes carry over.
Thanks,
-Alex
Hi,
This is a Apache Flex 4.13.0 release candidate 3. Please see the
RELEASE_NOTES and the README.
There is a known bug in Adobe Flash Builder 4.6 and 4.7 that causes a
issue when creating new applications.
Adobe has provided a work around for FB 4.7 but not for 4.6.
The release candidate can b
Hi,
> A - I would very much like to work in the 'develop' branches of the
> projects involved
Would be better if you work in a branch, if you;re worried about getting out of
date just merge develop into your branch every now and then. Git tools eg
source tree can make this easier for you.
> B
This is the discuss thread.
Changes from RC2 include:
-Install script handles AIR 14 beta
-Remove Zen sample theme as it was incomplete
-Fix for build numbers in flex-config. Looks like it was broken in 4.11
and onward
Thanks,
-Alex
Ok, just checking:
I create a local branch, let's call it 'vf2js', off 'origin/develop'. I
select this as my active branch. If I then 'rebase', all changes that have
been made to 'origin/develop' will be pulled into my local branch? So far,
so good. But now I want the world to see what's in my loc
>
> > B - In order for this to work, FalconJX needs to be part of the regular
> SDK
> > distribution.
> Why not make it an optional download via the installer ant script /
> installer?
>
Because that is just what this project needs, more complexity.
EdB
--
Ix Multimedia Software
Jan Luykenst
Hi,
> +
Not checked but should that be this instead like the the other changes?
Justin
Have a look at this ... I think it explains things quite well.
https://www.youtube.com/watch?v=cSf8cO0WB4o
Chris
Von: Erik de Bruin
Gesendet: Mittwoch, 2. Juli 2014 10:07
An: dev@flex.apache.org
Betreff: Re: New Flex to JS project
Ok, just checking:
I
Hi,
> - path="https://dist.apache.org/repos/dist/dev/flex/sdk/4.13.0/rc2/binaries/";
> file="apache-flex-sdk-4.13.0-bin" dev="true" nocache="true"/>
Any reason for nocache="true"? May make it a little hard to test caching.
Justin
On Wed, Jul 2, 2014 at 1:07 AM, Erik de Bruin wrote:
> Ok, just checking:
>
> I create a local branch, let's call it 'vf2js', off 'origin/develop'. I
> select this as my active branch.
So far so good.
> If I then 'rebase', all changes that have
> been made to 'origin/develop' will be pulled i
You don't loose the history when rebasing ... you sort of serialize it. Ok the
order of commits is messed up, but I never really care about this.
It might look less interesting in a GIT visualization tool where you could see
tons of different branches and merges, but it should be a lot easier to
On Wed, Jul 2, 2014 at 1:35 AM, Christofer Dutz
wrote:
> You don't loose the history when rebasing ... you sort of serialize it. Ok
> the order of commits is messed up, but I never really care about this.
> It might look less interesting in a GIT visualization tool where you could
> see tons of d
Hi Alex,
There is a bug I found in Flex SDK 4.13.0 RC2, I believe it's still in RC3.
http://apache-flex-development.247.n4.nabble.com/SkinnableTextBase-bug-found-in-Flex-4-13-0-RC2-td38785.html
I think we can quick fix it by adding a null check for skin.currentState in
that line:
https://iss
Merge sounds like the way to go... Let's not revisit the whole Git usage
thread(s) ;-)
EdB
On Wed, Jul 2, 2014 at 10:39 AM, OmPrakash Muppirala
wrote:
> On Wed, Jul 2, 2014 at 1:35 AM, Christofer Dutz >
> wrote:
>
> > You don't loose the history when rebasing ... you sort of serialize it.
>
If applicable please add the steps you take to the wiki so others can know
how to get started as well. IE, something like this for setting up your
environment,
https://cwiki.apache.org/confluence/display/FLEX/1.+Setting+up+the+SDK.
There may be something added since I last checked.
On Wed, Jul 2,
I know you hate this legal stuff, but don't forget your Apache headersŠ
-Alex
On 7/2/14 7:24 AM, "erikdebr...@apache.org" wrote:
>Repository: flex-asjs
>Updated Branches:
> refs/heads/VF2JS [created] 95b41cfca
>
>
>First component, more to serve as a placeholder than anything else ;-)
>
>Signe
As soon as I know what steps there are, sure ;-)
I'll create a guide to using this stuff. The idea is to keep it simple:
nothing special required on the MXML/AS side (business as usual, vanilla
Flex), with an FB 'External Tool' to launch the JS publishing. The
Installer should be able to handle mo
Hi,
> 1) two new namespaces and accompanying projects in the main Flex SDK:
> vf2js_mx and vf2js_s. These namespaces will contain shim objects for (you
> guessed it) MX and Spark components.
Perhaps a friendlier namesace name? mxshim and sshim? or just mxjs and sjs?
Justin
Wow, seriously? My blood pressure is still maxed out from the discussion on
private@, and this is how you decide to welcome this project? Come on, man,
the priorities of this project really, really need to change.
I have not released anything, and I am a long way away from doing that in
any case,
Nobody will ever have to use these namespaces in their projects. The
compiler replaces the 's' and 'mx' namespaces with 'vf2js_mx' and 'vf2js_s'
in a temp copy of the project which is only used published to JS before
being deleted again. The shim components in these namespaces are there only
to get
Sounds interesting. Is there any way to get more information on what
kinds of changes are going to be made? More comments in-line.
On 7/2/14 12:33 AM, "Erik de Bruin" wrote:
>Hi,
>
>I'm working on creating a way to publish vanilla Flex SDK projects to
>JavaScript on latest gen browsers. This p
On git: let's assume I work on these branches for 6 months (while keeping
them up to date with develop). How much of a hassle will it be to merge
them back into their respective 'develop' branches?
EdB
On Wed, Jul 2, 2014 at 10:39 AM, OmPrakash Muppirala
wrote:
> On Wed, Jul 2, 2014 at 1:35 A
>
> Sounds interesting. Is there any way to get more information on what
> kinds of changes are going to be made? More comments in-line.
>
I'm basically making this up as I go along, but... The workflow for
producing SWFs doesn't change at all. Hence 'Vanilla Flex'. If someone
wants to publish h
Hi Darkstone,
Are you saying this is a new problem introduced in RC2 or does it also
exist in 4.12.x? I don't see any changes of significance to that class in
4.13.
I'm really trying to get this release over with. There will be a 4.14 and
hopefully we can include a fix then.
Apache mailing lis
I think that can be removed.
On 7/2/14 1:15 AM, "Justin Mclean" wrote:
>Hi,
>
>> ->path="https://dist.apache.org/repos/dist/dev/flex/sdk/4.13.0/rc2/binaries
>>/" file="apache-flex-sdk-4.13.0-bin" dev="true" nocache="true"/>
>
>Any reason for nocache="true"? May make it a little h
On Jul 2, 2014 8:21 AM, "Erik de Bruin" wrote:
>
> On git: let's assume I work on these branches for 6 months (while keeping
> them up to date with develop). How much of a hassle will it be to merge
> them back into their respective 'develop' branches?
That depends on what files you touch and wha
That's kind of what I would've guessed. Still makes me wonder about auto
generating the shims from your JS source.
FWIW, FlexJS is eventually going to need to find a way to warn folks about
use of E4X and a few other things in a FlexJS app. I was going to try to
inject that intelligence into Fal
>
> I'm still trying to figure out why your Vanilla implementation, given it
> is going to have custom swcs, isn't just another FlexJS library. I think
> the main difference is in the reporting of missing mx/spark APIs. Is
> there more?
No, there is less. No special namespaces or MVC for the Fl
On 7/2/14 10:05 AM, "Erik de Bruin" wrote:
>>
>> I'm still trying to figure out why your Vanilla implementation, given it
>> is going to have custom swcs, isn't just another FlexJS library. I
>>think
>> the main difference is in the reporting of missing mx/spark APIs. Is
>> there more?
>
>
>N
My apologies. My memory of the policy was incorrect. Further reading of
the policy indicates that you can wait until release to correct the
headers.
-Alex
On 7/2/14 8:01 AM, "Erik de Bruin" wrote:
>Wow, seriously? My blood pressure is still maxed out from the discussion
>on
>private@, and thi
Hey everyone,
Trying to tackle one of the bug tickets, FLEX-34378. I've been able to
produce a patch file and attach to the ticket, but I also wanted to create
a pull request.
A while ago I forked the code and generated by first pull request
(FLEX-34324). It was nice and clean and the request onl
>
> The namespace and MVC in in the new AS/JS framework but I don't think
> there is anything in the current FlexJS setup what would block you from
> swapping in a set of SWCs that use the mx and/or spark NS.
>
No, I'm actually swapping OUT the mx and/or Spark NS from the regular SDK.
VF2JS projec
No worries. I'm a bit on edge lately, so I'm sorry for my harsh reply. But
I'm not kidding about them 'legal' reflexes... They need to be put on the
back burner for a while.
EdB
On Wed, Jul 2, 2014 at 8:23 PM, Alex Harui wrote:
> My apologies. My memory of the policy was incorrect. Further
It's pretty much one command. The same one you use to keep develop current
with your branch :)
-Nick
On Wed, Jul 2, 2014 at 11:20 AM, Erik de Bruin wrote:
> On git: let's assume I work on these branches for 6 months (while keeping
> them up to date with develop). How much of a hassle will it
Erik
Right now I'm on a stage of prove of concept using FlexJS and using Polymer
as the output template. I made with success at least with a simple button.
Polymer has similar architecture as Flex.
Is there a way I can share each other code? I'm not so familiar with
vanila sdk.
Best Regards
-
Okay, I got it now. Had to re-fork. Basically I incorrectly fetched from
apache/flex-sdk by creating another remote link. Should have used the
existing one to get from "upsteam".
Chris
On Wed, Jul 2, 2014 at 11:38 AM, Chris Martin wrote:
> Hey everyone,
>
> Trying to tackle one of the bug ti
Glad you figured it out. I was trying to follow what was going on --
because I've done similar and have not had an issue :)
-Nick
On Wed, Jul 2, 2014 at 5:01 PM, Chris Martin wrote:
> Okay, I got it now. Had to re-fork. Basically I incorrectly fetched from
> apache/flex-sdk by creating anot
yeah. How do you update your fork when you notice that develop has been
updated at apache/flex-sdk? I have the following remote entries for
flex-sdk.
C:\Users\cmartin\Documents\GitHub\flex-sdk>git remote -v
origin https://github.com/chrsmrtn-/flex-sdk.git (fetch)
origin https://github.com/chrsm
Hi all,
Accordingly to the open stats published at http://www.seethestats
.com/site/flex.apache.org "doc-getstarted.html" page got 19 155 hits, which
makes it second most popular
page at Apache Flex website.
But the content of this page was not updated quite for a long time now,
which lowers the
Thanks to Aaron Nash, we now have skins for Android 4.3 - TextArea and
TextInput. In the meantime, I have completed the Callout and related
skins.
I have updated the wiki page here:
https://cwiki.apache.org/confluence/display/FLEX/Progress+info+for+Android+4.x+skins
BusyIndicator is next on my p
+1
(although I would like to see that RSL numbering fixed on the next one)
-Mark
On Wed, Jul 2, 2014 at 4:06 AM, Alex Harui wrote:
> Hi,
>
> This is a Apache Flex 4.13.0 release candidate 3. Please see the
> RELEASE_NOTES and the README.
>
> There is a known bug in Adobe Flash Builder 4.6 a
Hi Alex,
This bug might probably exist in 4.12.x, but I never got the chance to find it
until 4.13.0 RC 2.
Well, I have workarounds to this bug, so no hurry, release 4.13.0 is the first
priority. I might fix the bug myself once I accept that invitation : )
Sent from DarkStone's iPhone
2014-07-
Thanks for bringing notice to this issue. Responses inline:
On Wed, Jul 2, 2014 at 4:02 PM, JabbyPanda wrote:
> Hi all,
>
> Accordingly to the open stats published at http://www.seethestats
> .com/site/flex.apache.org "doc-getstarted.html" page got 19 155 hits,
> which
> makes it second most po
Hi,
He been about for a while, low level active on the dev list, raised several
JIRAs, 3 with patches, the last patch even included a mustella test!
Thanks,
Justin
Well I totally stuffed that up didn't I :-)
Hi,
Please ignore this thread and please don't respond to it. It was meant to go to
private@.
Thanks,
Justin
HI,
May be just an Adobe / download issue but currently I'm getting this when
trying to compile. I've tried twice now. flex_sdk_4.6.0.23201B.zip is taking
15+ minutes to download.
download-adobe-flex-sdk:
[get] Getting:
http://fpdownload.adobe.com/pub/flex/sdk/builds/flex4.6/flex_sdk_4.6
Hi,
-1 binding (unable to use SDK to compile mobile or AIR apps and RSL issue)
Tested on OSX:
- signatures and md5 ok
- LICENSE, NOTICE, README, RELEAE_NOTES all good
- source header all good
- no binary files in source release
- can compile form source
- currently one mustella test failing but
Hi,
After several ties the Adobe SDK download worked.
However I'm concerned about the RSL issue, the flex config XML file has entries
like this in it:
libs/mx/mx.swc
mx_4.13.0.0.swf
But the RSLs have the current date:
ls frameworks/rsls/
Hi,
I notice there quite a bit of difference between resulting SDKs when installing
with installer and by the recommend scripts.
After sorting out the noise - an SDK produced by the installer is missing:
- Anything to do with producing asdocs this includes asdoc files in project sub
directorie
On Wed, Jul 2, 2014 at 9:48 PM, Justin Mclean
wrote:
> Hi,
>
> I notice there quite a bit of difference between resulting SDKs when
> installing with installer and by the recommend scripts.
>
> After sorting out the noise - an SDK produced by the installer is missing:
> - Anything to do with pro
Hi,
> Which ones are because of the Installer? We definitely want to fix them.
Hard to tell after the SDK has been produced but the lib jars at least need
double checking and while I hate to bring it up the licenses may be an issue.
> We can probably live with the differences caused by the bina
Hi All,
Is there any workaround for the RSL issues which Justin mentioned in this
thread..? Because our product heavily depends on RSL. This RSL issue will
definitely hold us to update our sdk from 4.9.1 to 4.13.x (if there is no
workaround)
*Regards,S. Jagan Langa* *|* *Project Manager*
*Syb
>I've been doing that as well for the last couple of years. I spent most of
>my time getting FlexJS off the ground. But I don't agree with some of the
>basic assumptions and choices that FlexJS has made. I said little or
>nothing about it to give FlexJS a proper head start and to keep you and
>P
Hi Igor,
Can you donate some of your FlexJS/Polymer code to Apache? We've already
got the beginnings of Cordova, Jquery and CreateJS. Polymer sounds like a
good addition.
Thanks,
-Alex
On 7/2/14 1:31 PM, "Igor Costa" wrote:
>Erik
>
>Right now I'm on a stage of prove of concept using FlexJS a
Hi,
> Is there any workaround for the RSL issues which Justin mentioned in this
> thread..?
There are several fairly easy workarounds:
- fix the the version number in the are and flex config files by hand
- copy over the correct config over the bad ones
- the release compile end ups up with the c
What steps did you take to set up an SDK with the recommended scripts.
Did you unpack the source kit, run ant, then a particular script? Or
unpack a binary kit? Or did you run the script in a repo branch?
IMO, lib/afe.jar and lib/aglj40.jar shouldn't be in an SDK. They should
be in lib/external
Thanks for fixing the copyright date, I keep forgetting to do that.
Don't know if folks noticed, but a user was complaining that installer 3.0
was hanging in the AIR unzip and was using >3GB of memory. I had him try
a custom installer that forced GC more often and then the installer
worked. Is t
Hi,
> What steps did you take to set up an SDK with the recommended scripts.
Described in my vote:
ant
ant frameworks-rsls
./ide/constructFlexForIDE.sh .
> Did you unpack the source kit, run ant, then a particular script?
Run on the source kit.
> IMO, lib/afe.jar and lib/aglj40.jar shouldn't b
Is that for the flex-config.xml in ide/flashbuilder/config? I'm not
seeing that for the flex-config in the frameworks folder of the source or
binary kits.
It looks like it has been this way for a while?
The scripts might need help, but I think it is time to ship and deal with
it later. The vast
67 matches
Mail list logo
