On Sun, Aug 7, 2016 at 7:09 AM, Harbs wrote:
> I also noticed that setting the className of a UIBase sets the class of
> the element, but setting the id does not set the element id. That does not
> seem right to me. Any reason to not add the id to the element as well?
>
In MXML, the "id" propert
You can use React in TypeScript with the type definitions here:
https://github.com/DefinitelyTyped/DefinitelyTyped/tree/master/react
My team is using React in TSX which references AS libraries typed with
as2dts. For my work, the major advantage AS has over TS is the is/as
operators which can check
On Wed, May 25, 2016 at 9:44 PM, lizhi wrote:
> yes,i will try.but i do not know.why no debug theyself of flexjs sdk
> developer?
> why not do not like check out the spriteflexjs code.and debug.
>
> it is a sdk bug,look at the bug,will improte the flexjs sdk.
>
> why not it just useful for myself
The way I would debug this is to open Chrome dev tools and enable "Pause on
exceptions" using the button circled here:
http://image.prntscr.com/image/dcc65ace5dd146718a7c608619b6521d.png
Just refresh the page after enabling that option and Chrome will show you
the stack trace of the error and allo
I think Alex is right that the problem is your BitmapData.as constructor
calls super() yet the BitmapData class does not extend anything. That
would explain the TypeError on line 2367 because parentCtor would be
undefined.
Try removing this line:
https://github.com/matrix3d/spriteflexjs/blob/mast
Hi lizhi,
Most of us read the forum via e-mail and the Nabble forum website does not
forward attached images through e-mail, so we did not realize you had
shared a screenshot.
>From the screenshot, the errors are:
Uncaught TypeError: Cannot read property 'prototype' of undefined.
base.js:2367
Unca
Copying and pasting error messages may help. Screenshots may help as well.
Andy
On May 21, 2016 5:44 PM, "lizhi" wrote:
>
> i want push more details.
> but i do not speak english good.
> so i can not get the details.
> i am sorry.
>
.
omething like this:
>
> function asInt(value){
> var intVal:int = int(value);//this should call parseInt()
> var numVal:Number = parseFloat(value)
> return intVal == numVal ? intVal : 0;
> }
>
> On May 18, 2016, at 1:20 PM, Andy Dufilie wrote:
>
> > On May 1
On May 18, 2016 2:50 AM, "Harbs" wrote:
>
> Isn’t this a behavior change?
>
> What happens if you do "5.3 as int” in ActionScript? Don’t you get 5?
>
I just tested this and got null.
There is no "int" type in JavaScript, so I think the compiler should just
replace int with Number.
In general I would not recommend using "as Number" or "as int" anyway,
since it returns null if the value is of a different type, and null cast to
Number or int becomes zero rather than NaN.
On Mon,
On May 9, 2016 1:27 PM, "Alex Harui" wrote:
>
>
> I would not be in favor of actually copying constants since it would add
> to bloat. I'd rather work on inlining constants and looking up the chain
> at compile-time to find values to inline.
>
Inlining constants would only work correctly for str
You're right, it is possible, but I think the key word there was "easily."
:)
Sent from my Android
On May 9, 2016 23:59, "Alex Harui" wrote:
>
>
> On 5/9/16, 1:22 PM, "Josh Tynjala" wrote:
>
> >That's a good call, Andy. You're absolutely right that a for-each loop
> >could also have its own ret
There can also be return statements within loops as well as "break to
label" or "continue to label". It all seems very messy to me and I still
think it should be behind an optimization option. The current
implementation works perfectly and I would prefer the JS output to be
minimally different fr
If you do make this change it should be an optional compiler argument,
because this changes the behavior of the code. The following loop should
print 1 and 2, not undefined:
var array = [];
array[2] = 1;
array[4] = 2;
for each (var x in array)
trace(x);
On May 7, 2016 10:18 AM, "Alex Harui"
The Mozilla StringView project is LGPL 3.0 so we would have to ask if they
can change the license or swap it out with something else.
On Fri, Apr 15, 2016 at 2:27 PM, Andy Dufilie
wrote:
> I'm fine with that. It also uses StringView by Mozilla, ported to AS from
> JS:
>
> h
wrote:
>
>
> On 4/15/16, 6:57 AM, "Andy Dufilie" wrote:
>
> >I've ported a JavaScript ByteArray implementation to ActionScript and
> >fixed
> >several bugs here:
> >
> https://github.com/WeaveTeam/Weave/blob/develop/WeaveJS/src/weavejs/util/J
I've ported a JavaScript ByteArray implementation to ActionScript and fixed
several bugs here:
https://github.com/WeaveTeam/Weave/blob/develop/WeaveJS/src/weavejs/util/JSByteArray.as
(original: https://github.com/jamesward/JSAMF/blob/master/web/web/amf.js)
It does not have any write functionality
On Tue, Mar 15, 2016 at 8:00 PM, Alex Harui wrote:
>
> I just took a quick look at React. What is it you like about it? If it
> is the actual UI library itself, I would have to wonder what it would take
> to wrap up their UI Library so it could be used in MXML. The JSX I saw
> seemed to have a
That feature sounds fine to me.
On Thu, Mar 17, 2016 at 8:58 PM, Alex Harui wrote:
> Hi,
>
> As I've been trying to get FlexJS to compile Flex code, there are some
> pain points. One is the parent property of UI widgets.
>
> In IUIBase.as, we want the parent to be another IUIBase
>
> functi
The latest browsers do not support the octal case anymore so I'm inclined
to say it should be left alone. I doubt it would ever be an issue.
On Tue, Mar 15, 2016 at 7:55 PM, Alex Harui wrote:
>
>
> On 3/15/16, 3:57 PM, "omup...@gmail.com on behalf of OmPrakash Muppirala"
> wrote:
>
> >I think
Mar 15, 2016 at 4:52 PM, Alex Harui wrote:
>
>
> On 3/15/16, 1:49 PM, "Alex Harui" wrote:
>
> >
> >
> >On 3/15/16, 1:35 PM, "Andy Dufilie" wrote:
> >
> >>It's not a good idea to make built-in functions behave differently in JS
This is great - I didn't know dts2as [1] was a thing.
I recently forked as3-to-typescript [2] which uses code ported to
TypeScript from FlexPMD [3], fixed all the bugs I could find, and made it
output just definitions rather than trying to generate all the AS code in
TS syntax. After hearing about
It's not a good idea to make built-in functions behave differently in JS
versus AS.
It's not a required param in AS, and this change makes the following
evaluate to true in AS but false in the cross-compiled JS:
Number("0xFF00FF") == parseInt("0xFF00FF")
On Tue, Mar 15, 2016 at 3:22 PM, Alex H
On Feb 23, 2016 19:34, "Alex Harui" wrote:
> OK, so I'll get the compiler to map Date's properties to getX/setX calls.
>
> For Array.sort, I guess we should map it to some other function in some
> other class. Should we generalize such a mapping or is that the only one?
>
> -Alex
>
>
These cons
aightforward but anything not as obvious just get flagged.
> >
> >My 2 cents,
> >Peter Ent
> >Adobe Systems/Apache Flex Project
> >
> >On 2/23/16, 1:44 PM, "Andy Dufilie" wrote:
> >
> >>AS has getHours() and setHours() too. If the code is upda
AS has getHours() and setHours() too. If the code is updated to use those
it should work.
Date is not the only class that has differences - Array.sort() is not the
same in AS and JS.
I think the most the compiler should do is provide a warning when using
known AS features that do not exist in JS,
Untyped variables are undefined by default. ActionScript does automatic
coercion when a variable is typed, so it becomes null for String or Object,
NaN for Number, or 0 for uint/int. Currently, this automatic coercion is
lost during cross-compilation.
On Wed, Feb 10, 2016 at 4:30 PM, Harbs wrote
The insert code looks fine to me, assuming children of XML are stored as
properties "0", "1", "2", ...
It's shifting all children with index >= i and saving P starting at child
index i.
[[Put]] is overwriting a child at a given index, and
insertChildBefore/insertChildAfter/prependChild should be d
On Wed, Jan 20, 2016 at 1:35 PM, Alex Harui wrote:
> The FlashPlayerDebugger can be downloaded from
> http://www.adobe.com/support/flashplayer/downloads.html
> (it says so in our READme ;-))
>
>
That's where I get the debugger for the browser, but I wasn't sure
specifically where FlashPlayerDebug
I've been struggling for a while trying to find the right settings that
will let me build FlexJS from the git repositories on Windows. I've
documented my steps here now that I've managed to get the
flex-asjs/build.xml/binary-release ant target to work:
https://github.com/adufilie/flex-dev-setup.
I
On Tue, Jan 19, 2016 at 2:39 PM, Alex Harui wrote:
> IMO, a base class shouldn't know about any extensions.
>
I agree, and I don't think we should bend over backwards trying to support
this pattern.
In the latest version, FalconJX does not remove circular dependencies by
default. You must specify -remove-circulars when running the build command.
On Jan 16, 2016 04:55, "lizhi" wrote:
> package
> {
> /**
> * ...
> * @author lizhi
> */
> public class A
file and method name in the JIRA issue. (if this is Falcon)
>
> Mike
>
> On Fri, Jan 15, 2016 at 9:55 AM, Andy Dufilie
> wrote:
>
> > I would like to write some unit tests for issues I know about, but I'm
> > unsure how to proceed. Should I commit the unit tes
I would like to write some unit tests for issues I know about, but I'm
unsure how to proceed. Should I commit the unit tests even if they
currently fail? I can create JIRA issues, but I know that unit tests are
helpful as well so I would like to know how to contribute those even if I
don't curren
Josh, does this commit resolve
https://issues.apache.org/jira/browse/FLEX-35004 ?
On Tue, Jan 12, 2016 at 11:24 AM, Alex Harui wrote:
> I haven't done thorough research, but I think this change is causing the
> last remaining failure in the Falcon build. The failing case is
> org.apache.flex.co
http://git-wip-us.apache.org/repos/asf/flex-sdk/diff/991f2c29
> >
> > Branch: refs/heads/develop
> > Commit: 991f2c294422da189c0f80f63829623895995da6
> > Parents: 386c8bb
> > Aut
Setting a tag seems like the best way.
git tag mytagname
git push origin mytagname
then do whatever you want.
On Mon, Jan 11, 2016 at 4:10 PM, OmPrakash Muppirala
wrote:
> Maybe this will work (based on this stackoverflow answer [1])
>
> 1. Move the code to its own branch.
>
> 2. Then tag th
The installer worked for me on Windows, though it took a long time to
download everything.
I didn't get any error, but the swfobject url should probably have
/{branch}.zip at the end:
https://github.com/swfobject/swfobject/archive/master.zip
or
https://github.com/swfobject/swfobject/archive/2.2.zi
On Sun, Jan 10, 2016 at 4:53 PM, Harbs wrote:
> I’m trying to figure out namespace behavior.
>
> I’m probably tired, but can someone tell me why the “catalog_item” var in
> this test (on line 70) does not get any results?
>
> https://gist.github.com/Harbs/c24285b3af80eeed251d
>
I put your code
; Mike
>
> On Sun, Jan 10, 2016 at 10:40 AM, Harbs wrote:
>
> > Welcome Andy!
> >
> > It great to have you on board!
> >
> > Harbs
> >
> > On Jan 10, 2016, at 5:39 PM, Alex Harui wrote:
> >
> > > Hi folks,
> > >
> > > The
I understand what you're trying to do, and I want the same things. The
greatest thing about FlexJS is that I am able to easily port my large
project (seven years of development work) with minimal change. However,
since my project relies heavily on the exact behavior of every AS language
feature,
On Fri, Jan 8, 2016 at 12:53 PM, Alex Harui wrote:
>
> On 1/8/16, 7:23 AM, "Andy Dufilie" wrote:
>
> >In all above cases, if Foo is a primitive type, function-style casting can
> >never be optimized out because Foo(x) will actually change the type of x.
> &g
On Fri, Jan 8, 2016 at 1:43 AM, Alex Harui wrote:
>
> But, IMO, for FlexJS, what really matters is what should happen in the JS
> output, and the essence of this thread is that in many cases, we can
> optimize away the call to Language.as.
>
> Right now, "function" casting also results in a Lan
On Thu, Jan 7, 2016 at 10:54 PM, Josh Tynjala wrote:
> That's odd. I swear I remember someone from Adobe once explaining that
> function-style casting is faster than as-style casting (with the exception
> of primitive types that have a top level function that makes function-style
> casting imposs
On Thu, Jan 7, 2016 at 8:32 PM, Alex Harui wrote:
>
> On 1/7/16, 5:24 PM, "Andy Dufilie" wrote:
> >if (obj is Thing1)
> >(obj as Thing1).foo(a,b);
> >else if (obj is Thing2)
> >(obj as Thing2).bar(x,y);
>
> There are no plans to change "
On Thu, Jan 7, 2016 at 4:55 PM, Alex Harui wrote:
> How many of you use the "as" keyword as part of a test?
I have a huge code base and I use "as" / "is" everywhere (
https://github.com/WeaveTeam/Weave). My code depends on it behaving the
same way it does in ActionScript. IMO, being able to ea
button is null because document.createElement("button") is not an instance
of MyButton.
Is this supposed to work?
MyButton.as
package {
public class MyButton extends HTMLButtonElement {
public function MyButton() {
super();
this.addEventListener("click", button_clickListener, false);
}
private fun
On Tue, Jan 5, 2016 at 9:37 AM, Harbs wrote:
> That’s not going to work if you need to implement toXMLString().
>
> Imagine an mxml file where all the CDATA becomes escaped text. It’s not
> going to work very well… ;-) The same goes for lots of other XML file uses.
>
> Apparently this recommendat
On Tue, Jan 5, 2016 at 7:10 AM, Harbs wrote:
> I cannot find a recommended way to handle CDATA in the DOM4 spec.[3]
>
> Does anyone have a recommendation o how to handle this?
>
In place of a CDATA text section you can just escape the text using xml
escaping.
On Thu, Dec 31, 2015 at 2:42 AM, Alex Harui wrote:
>
> On 12/30/15, 10:31 PM, "Andy Dufilie" wrote:
> >- When I run flexjs/js/bin/compc I get a bin-release folder with the JS
> >output files. I then try to run closure manually on these files, and I
> >get
&g
On Mon, Jan 4, 2016 at 11:17 AM, Harbs wrote:
> For literals, yes, we’d for the most part just pass the whole thing as a
> string to the constructor. The only exception I can think of is bracket
> notation:
>
> var foo:XML = foo;
> would have to become:
> var foo:XML = new XML(' attr="‘+myAttr+’"
On Fri, Jan 1, 2016 at 3:28 AM, Alex Harui wrote:
> The externc.createjs.swc target calls another ant script that munges the
> externs. You could try doing that temporarily for the externc.js.swc
> target.
>
>
Oops, I was looking at the wrong target. I see the externc.js.swc target
now - it comp
On Thu, Dec 31, 2015 at 11:31 PM, Andy Dufilie
wrote:
> I've found this page which explains how to compile newer Java code to run
> under earlier versions (write Java 1.7 code, run under 1.6):
>
> http://javarevisited.blogspot.com/2015/05/fixing-unsupported-majorminor-version
Alex Harui wrote
> One thing I’ve hit: FB is apparently stuck using Java 1.6. That’s how
> old the version of Eclipse is that it uses. Some of the code you’ve
> written uses Java 1.7. Can we change back to Java 1.6 syntax? Really,
> only the code paths used by FB seem to need to be in Java 1.6
On Thu, Dec 31, 2015 at 5:41 PM, Andy Dufilie
wrote:
> Now that I can compile with js.swc, I am getting a lot of errors because
> the type information for many JS functions is incorrect. For example, it
> says Array's sort() is a void function. Where is the source code for js.swc
&
On Thu, Dec 31, 2015 at 12:17 PM, Josh Tynjala
wrote:
> You might try using asjsc instead of mxmlc if you want to compile with
> JS.swc.
>
> - Josh
>
I had tried that before through Flash Builder but it was giving me an error
("Unsupported major.minor version 51.0"). I looked at it again and fo
On Wed, Dec 30, 2015 at 1:05 PM, Alex Harui wrote:
> Try: -js-compiler-option="--compilation_level WHITESPACE_ONLY"
That worked after fixing the special characters. Is there a place that
options like this are documented?
Now that I'm compiling via flexjs/js/bin/mxmlc using the WHITESPACE_ONLY
57 matches
Mail list logo
