On Wed, Sep 25, 2013 at 8:41 PM, Alex Harui aha...@adobe.com wrote:
Turns out there is a bug in the Falcon code that the search for IFlexInfo
only is in the baseclass and not deeper.
JQuery's app does not directly implement IFlexInfo. Adding that is
probably the cheapest workaround for you.
Thank you Alex,
I will check it now.
Because in the 4.1.0 the tree is working without problem but after updated it
throws that error.
Thanks for the tip now i know where to look.
Miguel
From: aha...@adobe.com
To: dev@flex.apache.org
Date: Wed, 25 Sep 2013 09:15:42 -0700
Subject: Re: tree
On 25/09/2013 20:55, aYo ~ wrote:
In developing mxml classes using code behind I've often accessed protected
class variables via the mxml attributes. I believe it's allowed
Yeah, me too - the MXML is extending the base class, so of course gets
access to package-local variables.
Tom
I use...
[anypath]/flex-sdk
[anypath]/flex-tlf
Just the default folder naming. Fred can tell you how to rename/move change
things to fully customize the pathing layout.
-Mark
-Original Message-
From: Maurice Amsellem [mailto:maurice.amsel...@systar.com]
Sent: Thursday, September
Thanks Fred,
I will continue to post questions on the ML for others.
Please ignore them until this week-end :-).
Maurice
-Message d'origine-
De : Frédéric THOMAS [mailto:webdoubl...@hotmail.com]
Envoyé : jeudi 26 septembre 2013 14:24
À : dev@flex.apache.org
Objet : RE: Building the
Yes, I am on the develop branch.
I have also read the basic docs on git since I am new to git ( they are very
well written).
the wiki will tell you how to configure git
You mean https://cwiki.apache.org/confluence/display/FLEX/Apache+Flex+Wiki; ?
OK, I will read it.
Maurice
-Message
I have started reading the wiki on Setting up Manually, and it says :
TLF: git clone [https://git-wip-us.apache.org/repos/asf/flex-tlf.git] flex-tlf
SDK: git clone [https://git-wip-us.apache.org/repos/asf/flex-sdk.git] flex-sdk
However, http://flex.apache.org/dev-sourcecode.html says:
git clone
https://git-wip-us.apache.org/#contents
-Message d'origine-
De : Maurice Amsellem [mailto:maurice.amsel...@systar.com]
Envoyé : jeudi 26 septembre 2013 14:31
À : dev@flex.apache.org
Objet : RE: Building the SDK from git
Yes, I am on the develop branch.
I have also read the basic docs
My .gitconfig has only two lines, name and email. :-(
I will read https://git-wip-us.apache.org/#contents to set the correct options.
Thanks
Maurice
-Message d'origine-
De : Frédéric THOMAS [mailto:webdoubl...@hotmail.com]
Envoyé : jeudi 26 septembre 2013 14:45
À :
Hi,
With TLF environment variables work fine.
On OSX I have the following:
TLF_HOME=/Users/justinmclean/Documents/ApacheFlexTLFGit/
AIR_HOME=/Users/justinmclean/Documents/air/mac/3.1/
PLAYERGLOBAL_HOME=/Users/justinmclean/Documents/player/
Hi,
Which source is the most up-to-date ?
You want the develop branches of both the SDK and TLF.
Thanks,
Justin
On 9/25/13 11:57 PM, OmPrakash Muppirala bigosma...@gmail.com wrote:
On Wed, Sep 25, 2013 at 8:41 PM, Alex Harui aha...@adobe.com wrote:
Turns out there is a bug in the Falcon code that the search for
IFlexInfo
only is in the baseclass and not deeper.
JQuery's app does not directly
Also, I noticed .actionScriptProperties and .flexLibProperties are missing from
framework/projects/experimental
Not committed or is it on purpose ?
Maurice
-Message d'origine-
De : Justin Mclean [mailto:jus...@classsoftware.com]
Envoyé : jeudi 26 septembre 2013 15:12
À :
Understood.
So maybe they should be included in the .gitignore (along with .project, etc.) ?
Maurice
-Message d'origine-
De : Erik de Bruin [mailto:e...@ixsoftware.nl]
Envoyé : jeudi 26 septembre 2013 15:45
À : dev@flex.apache.org
Objet : Re: Building the SDK from git
On purpose, we
Yes, I have tried to keep a journal of what I do and the issues I am facing.
I will do it with that in mind from now on. (ie updating the resources at the
end)
Maurice
-Message d'origine-
De : Erik de Bruin [mailto:e...@ixsoftware.nl]
Envoyé : jeudi 26 septembre 2013 15:56
À :
Hi,
So maybe they should be included in the .gitignore (along with .project,
etc.) ?
-1 to that. Anything that is actually in version control should not be in
.gitignore, this has caused far too many issues in the past, including breaking
a release candidate.
IMO if something in version
Hi,
Yes, I think that is the idea... I think we somehow all ended up
with it in our local ignores
It's not in mine and I've not seen any issues, but they probably shouldn't be
in there in the first place, given ant is the preferred build method.
Thanks,
Justin
Hi,
My config is the following:
- I have created a project in FB to map projects/experimental = added
.actionScriptProps and .flexLibProps
- Added both entries to .gitignore, to make sure they will not end up if any
commit
given ant is the preferred build method.
Please can you share how you
Hi,
Please can you share how you proceed to compile when you make even small
changes to the code and would like to test them.
Are you using an IDE at all ?
I normally use Flash Builder to edit files (but sometimes vi from the command
line) and I normally compile using ant from the command
Hi,
However playerglobal.version is still 11.1. Does this mean I should compile
agains FP 11.1?
You can compile against any version of the Flash Player from 10.2 onwards, we
try and keep it backwards compatible. People who use AIR generally want the
latest, people who use Flash Player may
Thanks, that's really helpful.
I understand I launch the general build.xml the first time only, and then only
need to run the build.xml for the project I am working on (eg.
experimental/build.xml) which will compile rather quickly.
Is that correct ?
Maurice
-Message d'origine-
De :
BUILD SUCCESSFUL!
YEEHAA!
Do you me to recap the few issues I have faced as a newbie?
Maurice
-Message d'origine-
De : Maurice Amsellem [mailto:maurice.amsel...@systar.com]
Envoyé : jeudi 26 septembre 2013 17:42
À : dev@flex.apache.org
Objet : RE: Building the SDK from git
Thanks,
Hi Maurice,
The grant was filed today. Please make a snapshot of the code available
when you can. You can email the link or source to me directly. Any other
PMC'rs want to help with the IP review?
-Alex
On 9/25/13 7:56 PM, Alex Harui aha...@adobe.com wrote:
Again, none of this is required
On 9/26/13 9:12 AM, Maurice Amsellem maurice.amsel...@systar.com wrote:
BUILD SUCCESSFUL!
YEEHAA!
Do you me to recap the few issues I have faced as a newbie?
Sure. I think you'll be getting write access soon, so then you'll be able
to update the README yourself.
-Alex
We also received the acknowledgments. That's good news.
Do I need to make the changes requested before (Apache headers, etc..) before
sending you the sources ?
Maurice
-Message d'origine-
De : Alex Harui [mailto:aha...@adobe.com]
Envoyé : jeudi 26 septembre 2013 18:31
À :
In my opinion the default appearance of Spark components in Flex could be
greatly improved.
As an example, the Ext JS (AKA Sencha) javascript framework has many
components very similar to the ones in Flex, indeed Ext has billed itself as
a Flex replacement. Ext has a gray theme (amongst others)
In theory, Systar's legal department should be requiring some form of
header on any code leaving your company. So yes, adding headers is
advised and it would be my preference and probably save time in the end if
you use the Apache headers and NOTICE file.
IMO, no other changes to the code are
Would be interesting to do. But I have a funny feeling there will be lots of
opinions on this area. Not to mention if you did rewrite the skins, would you
create them more in AS instead of mxml.
-Mark
-Original Message-
From: garethdanielsmith [mailto:garethdanielsm...@gmail.com]
River Song would say: Spoilers! The bane of time travellers all over
the universe...
;-)
EdB
On Thu, Sep 26, 2013 at 6:32 PM, Alex Harui aha...@adobe.com wrote:
On 9/26/13 9:12 AM, Maurice Amsellem maurice.amsel...@systar.com wrote:
BUILD SUCCESSFUL!
YEEHAA!
Do you me to recap the few
OK, I will add the Apache headers and Notice, and send you the code, tonight or
tomorrow latest...
Maurice
-Message d'origine-
De : Alex Harui [mailto:aha...@adobe.com]
Envoyé : jeudi 26 septembre 2013 18:43
À : dev@flex.apache.org
Objet : Re: [VOTE][LAZY] Accept donation of Systar
-Message d'origine-
De : Maurice Amsellem [mailto:maurice.amsel...@systar.com]
Envoyé : jeudi 26 septembre 2013 18:58
À : dev@flex.apache.org
Objet : RE: Building the SDK from git
Hi,
This is my (uncensored :-) feedback following the SDK build.
1) Doc inconsistency:
There are 3
It depends the OS, on Windows it is recommended to set it to true,
input on others but if you're on windows but use Cygwin, it has to be set
to input.
-Message d'origine-
De : Kessler CTR Mark J [mailto:mark.kessler@usmc.mil]
Envoyé : jeudi 26 septembre 2013 19:20
À :
Yeah I'm Windows with TortoiseGit, didn't have to change anything out of the
box for that. It's setup to translate into Windows cr/lf and then back again
automatically.
-Original Message-
From: Frédéric THOMAS [mailto:webdoubl...@hotmail.com]
It depends the OS, on Windows it is
At least it what I deducted because it happened to me I checkout the dev
branch with autocrlf=true and wasn't able to build from ant on Cygwin, the
build scripts wasn't read correctly, changing it by autocrlf=input, checking
out again made it running correctly.
-Message d'origine-
De :
Yep, I guess it is not the same doing a check out from windows (whatever the
tool) and from Cygwin.
-Message d'origine-
De : Kessler CTR Mark J [mailto:mark.kessler@usmc.mil]
Envoyé : jeudi 26 septembre 2013 19:29
À : dev@flex.apache.org
Objet : RE: Building the SDK from git
Yeah
I am adding the headers now.
I noticed some of the components in experimental had mention of the original
author, in addition to Apache Header. (see below).
Should I do the same ?
Eg. @author Maurice Amsellem for SYSTAR (maurice.amsel...@systar.com).
Hi Maurice,
Most projects at Apache discourage author tags. There is no official
policy prohibiting and Apache Flex has not made any decision.
The general reasoning is that there are currently 40+ committers and
growing on Apache Flex and if everyone who touched a file added an author
tag,
On Thu, Sep 26, 2013 at 9:36 AM, garethdanielsmith
garethdanielsm...@gmail.com wrote:
In my opinion the default appearance of Spark components in Flex could be
greatly improved.
As an example, the Ext JS (AKA Sencha) javascript framework has many
components very similar to the ones in Flex,
Thanks for the clarification.
I won't put the @author tag then.
Maurice
-Message d'origine-
De : Alex Harui [mailto:aha...@adobe.com]
Envoyé : jeudi 26 septembre 2013 19:45
À : dev@flex.apache.org
Objet : Re: [VOTE][LAZY] Accept donation of Systar DataGrid for Mobile
Applications
Hi
Darrell, Gordon (mostly),
I'm having trouble with the codegen for a Binding destination function. I
have an MXMLPropertySpecifierNode for the property to be set. This
property can be protected or private as well as public. I want to use the
OP_setproperty instruction which seems to take a
I don't think using a Binding should be necessary. I think the bug is the
getName() of MXMLPropertySpecifierNode isn't returning the right kind of
Name when the property is protected or private.
- Gordon
On 9/26/13 11:13 AM, Alex Harui aha...@adobe.com wrote:
Darrell, Gordon (mostly),
I'm
It turns out that MXMLPropertySpecifierNode just calls its definition's
getBaseName(). There doesn't seem to be any API on IDefinition to return
a fully qualified name. I can get the INamespaceReference from the
definition, but then how to I convert the INamespaceReference to a
Namespace I can
I'm in a meeting in San Jose and don't have access to the Falcon code at
the moment. But I think Idefinition (or maybe only DefinitionBase) has a
getMName() method that returns a Name.
- Gordon
On 9/26/13 11:37 AM, Alex Harui aha...@adobe.com wrote:
It turns out that MXMLPropertySpecifierNode
I was able to browse the Apache repo. getMName() is on DefinitionBase, but
not on IDefinition. So try
((DefinitionBase)propertyNode.getDefinition()).getMName() and see whether
it returns the right kind of Name when the property is private or
protected.
- Gordon
On 9/26/13 12:05 PM, Gordon Smith
I have sent some comments on the code earlier on this thread (jeu. 26/09/2013
02:15 (UTC+1)) , that could be helpful for the review.
Maurice
-Message d'origine-
De : Alex Harui [mailto:aha...@adobe.com]
Envoyé : jeudi 26 septembre 2013 21:00
À : dev@flex.apache.org
Objet : Re:
getMName returns a Name with the same namespace as calling
getNamespaceReference() on the definition.
That still doesn't help me when the property is a [Bindable] var in a
class in a SWC. The definition is a VariableDefinition with the
BindableNamespace which represents the backing variable for
Hi,
While preparing the potential integration of MobileDataGrid component to the
SDK, I ran into an unexpected issue.
It seems that MDG cannot be integrated into the experimental projects, because
it's a mobile component, and it relies on mobile-only classes ( such as
StyleableTextFiled or
I didn't work on Falcon's data bindingŠ Bruce Frazer implemented what's
there.
If you can get the synthetic SetterDefinition then you can get its Name
with getMName(). So find the code that synthesizes the getter/setter pair
for a [Bindable] var, and make it store references to the getter and
On 9/26/2013 4:38 PM, Maurice Amsellem wrote:
Does this mean we need to introduce a new mobileExperimental project ?
I didn't see any other responses, so I want to say that I like that
idea a lot.
--
Jeffry Houser
Technical Entrepreneur
http://www.jeffryhouser.com
203-379-0773
Yes, but it seems it's a lot of work, and has many impacts on the SDK to
introduce a new SWC.
(impacts global build files and config files, etc...).
Plus the multi-lingual resource bundles (although they would be empty for now).
Maurice
-Message d'origine-
De : Jeffry Houser
Op 26 sep. 2013 om 19:53 heeft OmPrakash Muppirala bigosma...@gmail.com het
volgende geschreven:
On Thu, Sep 26, 2013 at 9:36 AM, garethdanielsmith
garethdanielsm...@gmail.com wrote:
In my opinion the default appearance of Spark components in Flex could be
greatly improved.
As an
Hi,
In theory, Systar's legal department should be requiring some form of
header on any code leaving your company.
I'm not a laywer but my understanding was that code (or anyting) doesn't have
to have a copyright header in order to give protection to it's owners, that's
now automatic.
See
Hi,
It seems that MDG cannot be integrated into the experimental projects,
because it's a mobile component, and it relies on mobile-only classes ( such
as StyleableTextFiled or MobileSkin).
Which shouldn't be too hard to fix , I can help out there.
Thanks,
Justin
On 9/26/13 3:51 PM, Justin Mclean jus...@classsoftware.com wrote:
Hi,
Yes, but it seems it's a lot of work, and has many impacts on the SDK
to introduce a new SWC.
It would be more work but it's reasonable straight forward, again I can
help out there if needed.
How are you planning to help?
Hi,
How are you planning to help?
I fixed up build scripts and added a new projects before (eg apache) so can
help there if it's needed.
Do you think a new SWC is the right idea?
Either solution seems OK to me, not sure what the right one is.
Thanks,
Justin
StyleableTextField belongs mobilecomponents.swc
And MobileSkin belongs to components.swc.
Does this means you are going to add these two swc to the lib-path of
experimental.swc ?
Wouldn't that make it unusable in non-mobile apps ?
This is not an affirmation, just a question.
Maurice
Sorry, mobile.swc is a theme, so that's different.
Maurice
-Message d'origine-
De : Maurice Amsellem [mailto:maurice.amsel...@systar.com]
Envoyé : vendredi 27 septembre 2013 01:15
À : dev@flex.apache.org
Objet : RE: Mobile datagrid component integration
StyleableTextField belongs
This is not pertaining to just this case, but our strategy in general about
'experimental' swcs. If the intention is to warn the users that these
components have not been thoroughly tested, can we do that in a different
way? ie. README, Release Notes, ASDocs, custom metadata, etc.?
I think it
Hi,
This is not pertaining to just this case, but our strategy in general about
'experimental' swcs.
The original intent was a place for components that have not been fully tested,
been fully localised or not fully documented but in a good working state. It
was hoped by releasing them in
I figure it will be another day or so before we finish getting
write-access for Maurice. If you can get a new SWC together in that time,
that would be my preference. Jeffry liked it too.
-Alex
On 9/26/13 4:04 PM, Justin Mclean jus...@classsoftware.com wrote:
Hi,
How are you planning to
HI,
Done and checked into experimentalmobile branch, it will need a little bit of
work to make it compile but the structure and build files are there.
Thanks,
Justin
Excellent. Thanks.
On 9/26/13 5:43 PM, Justin Mclean jus...@classsoftware.com wrote:
HI,
Done and checked into experimentalmobile branch, it will need a little
bit of work to make it compile but the structure and build files are
there.
Thanks,
Justin
Hi Maurice,
I imported the two files into FB and I ran the Apache file scanner on the
folders. Looks good. Here are some things I saw:
1) If you plan to check in the FB project properties like
.actionScriptProperties, you will need to insert headers in them as well.
2) Are you planning to
On Sep 26, 2013 9:34 PM, Alex Harui aha...@adobe.com wrote:
Hi Maurice,
I imported the two files into FB and I ran the Apache file scanner on the
folders. Looks good. Here are some things I saw:
1) If you plan to check in the FB project properties like
.actionScriptProperties, you will
64 matches
Mail list logo
