Re: [LAST CALL] Release FlexJS/FalconJX 0.8.0

2017-04-26 Thread Alex Harui


On 4/26/17, 4:45 PM, "Justin Mclean"  wrote:

>Hi,
>
>> Not yet.  I will be seeking clarification on whether I can manually
>>create
>> the final result without it being called a derivative work.
>
>I assume you be asking this on the current legal discuss thread? Wouldn’t
>it be less effort to just do as Roy advises and add the MIT header?

I am only currently authorized to donate Adobe-owned stuff to Apache.  If
the externs turn out to be owned by CreateJS, I will have to delete them
from the repo, and some non-Adobe person will have to figure out what to
do, or I will have to go through an Adobe-side process of getting approved
to donate this to CreateJS, which is more hassle than just manually
creating the files.

>
>Even it you manually create the file it still may come down to an ASF
>policy question rather than a legal question which may also need to be
>resolved.

Well, we'll see what Roy says.

-Alex



Re: [LAST CALL] Release FlexJS/FalconJX 0.8.0

2017-04-26 Thread Justin Mclean
Hi,

> Not yet.  I will be seeking clarification on whether I can manually create
> the final result without it being called a derivative work.

I assume you be asking this on the current legal discuss thread? Wouldn’t it be 
less effort to just do as Roy advises and add the MIT header?

Even it you manually create the file it still may come down to an ASF policy 
question rather than a legal question which may also need to be resolved.

Thanks,
Justin



Re: [LAST CALL] Release FlexJS/FalconJX 0.8.0

2017-04-26 Thread Alex Harui


On 4/26/17, 3:51 PM, "Justin Mclean"  wrote:

>Hi,
>
>Alex give Roy’s last repose [1] on legal discuss are you OK for me to go
>ahead and add back in the MIT license?

Not yet.  I will be seeking clarification on whether I can manually create
the final result without it being called a derivative work.  Given that
Roy also seemed to say the patch files themselves shouldn't have ASF
headers, it might be most expedient for me to do away with the whole patch
process and manually create the final result.

-Alex




Re: [LAST CALL] Release FlexJS/FalconJX 0.8.0

2017-04-26 Thread Justin Mclean
Hi,

Alex give Roy’s last repose [1] on legal discuss are you OK for me to go ahead 
and add back in the MIT license?

Thanks,
Justin

1. 
https://lists.apache.org/thread.html/6506d8e5001b74f36b338a81b3175f94da0c79785af04c9cd06fa44c@%3Clegal-discuss.apache.org%3E

Re: [FlexJS][POC] JSFiddle for FlexJS / Compile as-a-service

2017-04-26 Thread Josh Tynjala
I think Microsoft's Monaco editor would be good for this sort of thing:

https://github.com/Microsoft/monaco-editor

It's the core editor used by Visual Studio Code, packaged up as a
standalone component.

I recently saw that a company called TypeFox found a way to make Monaco to
speak the "language server protocol" used by VSCode, with the language
server running on the backend instead of in a separate process on the
client.

http://typefox.io/teaching-the-language-server-protocol-to-microsofts-monaco-editor

Getting Monaco to work with the NextGenAS language server on the backend is
probably not trivial (and it definitely should wait for a future
iteration), but it would be really cool to see IntelliSense/completion
working from the browser.

- Josh

On Wed, Apr 26, 2017 at 2:55 AM, yishayw  wrote:

> I certainly don't think it's useless. I've been imagining something similar
> and it's great to see you take it up. I wonder what the effort would be to
> adapt Moonshine to be the web text editor...
>
>
>
> --
> View this message in context: http://apache-flex-
> development.247.n4.nabble.com/FlexJS-POC-JSFiddle-for-
> FlexJS-Compile-as-a-service-tp61369p61370.html
> Sent from the Apache Flex Development mailing list archive at Nabble.com.
>


Re: [FlexJS][POC] JSFiddle for FlexJS / Compile as-a-service

2017-04-26 Thread Josh Tynjala
For a while now, I've wanted to build something similar to TypeScript's
playground. I just haven't found the time yet.

https://www.typescriptlang.org/play/

Basically, one file/class in and one file out. The idea would be to keep it
super basic. Simply let people see the JavaScript that gets produced from
ActionScript and MXML with FlexJS. Maybe provide a couple of sample MXML/AS
files alongside an option to write your own code from scratch.

This could be a good MVP because it is something people will find useful,
but it can also serve as a foundation for a more complex CodePen-like
experience with multiple files in a simple project.

- Josh

On Apr 26, 2017 2:10 AM, "OK"  wrote:

Hi guys,
I remember that there were some discussions that it would be nice to have
something like JSFiddle or CodePen for FlexJS.
I also remember that Alex introduced a POC for live editing a couple of
months ago.
Cause I've no idea if something like this could be achieved as-a-service and
I'm honestly not smart enough to improve this approach I wonder if it would
be an option to just make the compiler available as-a-service and build an
app around it that consumes this service.

Although this approach is far away from live editing and you have to wait
until the compiler is done I  think that FlexJS would benefit a lot from
something like this:

There would be no more entry barrier to get in touch with FlexJS: Just call
the (web) app, code and compile it with one click.
Folks could share useful FlexJS snippets and with the ability to store this
stuff it could automatically grow to an archive of code snippets.
It would be very ergonomic to compile the code with different FlexJS
versions or compiler options.
At least you would have one browser window that contains the code and the
result so I think this could be also a little helpful tool during regular
app development.

So I started a little POC regarding the service:
I've just installed nodejs express and make Josh's asconfigc available via
an http endpoint.
This is just a few lines of code and it works as expected [1][2].
In the real world, the service would expect some FlexJS code and response
with the compiled code or an URL where the code lives.
Next step would be to build a FlexJS app that implements a CodePen inspired
UI and consumes the service.

Before starting with this I'd like to ask if you think that this approach is
worth following it or if you expect too many issues (maybe also legal
issues) or if you think it's completely useless.

Of course, it's a long way to making it happen and a powerful backend is
needed and this is probably not for free. I'm not yet familiar enough with
the various cloud services out there but I could imagine that something like
'AWS lamda' would work. Maybe Apache Infra provide something like this?

Thanks,
Olaf

[1] https://snag.gy/Bb83sq.jpg
[2] https://snag.gy/crM43o.jpg



--
View this message in context: http://apache-flex-development
.247.n4.nabble.com/FlexJS-POC-JSFiddle-for-FlexJS-
Compile-as-a-service-tp61369.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.


Re: [FlexJS] Layouts

2017-04-26 Thread Peter Ent
Let's do this then, after the merge of dual into develop:

AbsolutePositioningViewBeadBase will be the view bead to use for controls.
On the HTML platform, this bead will set its strand's style position to
"relative" if it is not set to either "absolute" or "relative" already. In
addition, this bead will have a setAbsolutePosition(child, x, y) function
that on Flash, just sets x and y while on HTML, sets left and top styles
as well as setting position:absolute style.

You could use it like this: Create a view bead class that extends this
class and set your custom view bead class in CSS for your control or add
it programmatically. The super.strand setter will take care of setting the
strand's position style. After you have created any elements required for
your control, position them using setAbsolutePosition() and do so in
response to changes in your control's size.

—peter



On 4/25/17, 4:47 PM, "Alex Harui"  wrote:

>
>
>On 4/25/17, 12:35 PM, "Peter Ent"  wrote:
>
>>Perhaps I am not understanding this.
>>
>>You create a control component extending UIBase. You create your view
>>bead
>>extending AbsolutePositioningViewBase. This bead makes sure that the
>>strand/host has position:relative set.
>>
>>So how do you set position:absolute on each of the control's parts?
>>Setting part.x and part.y won't be enough. You still need set
>>part.style.position = "absolute". I don't think that will translate over
>>to the HTML side and will not be useful to Flash.
>
>Well, you are right that there is more to it than just the ViewBase.
>
>However, we are in the business of encapsulating patterns.  We could do
>any combination of the following:
>
>1) have setter for x and y set position="absolute" but it won't change the
>parent's position style.  AbsolutePositioningViewBase would still do that.
>2) add a setChildPosition method on AbsolutePositioningViewBase.
>3) add xChanged,yChanged listeners to children
>4) add childAdded listener to parent and check the child's x,y value
>
>Thoughts?
>-Alex
>
>>
>>‹peter
>>
>>On 4/25/17, 2:08 PM, "Alex Harui"  wrote:
>>
>>>
>>>
>>>On 4/25/17, 10:55 AM, "yishayw"  wrote:
>>>
App devs may want to write custom controls and would expect flash and
js
behaviour to be the same. I think we should strive to eliminate use of
conditional compilation for app devs. If that's so we don't want them
changing an HTML only style (position).
>>>
>>>Agreed, which is why we'd offer different base classes for views, one of
>>>which assigns the position style.
>>>
>>>-Alex
>>>
>>
>



Re: [FlexJS][POC] JSFiddle for FlexJS / Compile as-a-service

2017-04-26 Thread Justin Mclean
Hi,

> There would be no more entry barrier to get in touch with FlexJS: Just call
> the (web) app, code and compile it with one click.

Great idea!

> Before starting with this I'd like to ask if you think that this approach is
> worth following it or if you expect too many issues (maybe also legal
> issues) or if you think it's completely useless.

No major legal issues that I’m aware of that will trip you up. As long as you 
respect Apache’s trademarks (which is easily done by adding a disclaimer in a 
footer) and select a domain name that’s not likely to cause an issue.

Thanks,
Justin

Re: [FlexJS][POC] JSFiddle for FlexJS / Compile as-a-service

2017-04-26 Thread yishayw
I certainly don't think it's useless. I've been imagining something similar
and it's great to see you take it up. I wonder what the effort would be to
adapt Moonshine to be the web text editor...



--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-POC-JSFiddle-for-FlexJS-Compile-as-a-service-tp61369p61370.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.


[FlexJS][POC] JSFiddle for FlexJS / Compile as-a-service

2017-04-26 Thread OK
Hi guys,
I remember that there were some discussions that it would be nice to have
something like JSFiddle or CodePen for FlexJS.
I also remember that Alex introduced a POC for live editing a couple of
months ago.
Cause I've no idea if something like this could be achieved as-a-service and
I'm honestly not smart enough to improve this approach I wonder if it would
be an option to just make the compiler available as-a-service and build an
app around it that consumes this service.

Although this approach is far away from live editing and you have to wait
until the compiler is done I  think that FlexJS would benefit a lot from
something like this:

There would be no more entry barrier to get in touch with FlexJS: Just call
the (web) app, code and compile it with one click.
Folks could share useful FlexJS snippets and with the ability to store this
stuff it could automatically grow to an archive of code snippets.
It would be very ergonomic to compile the code with different FlexJS
versions or compiler options.
At least you would have one browser window that contains the code and the
result so I think this could be also a little helpful tool during regular
app development.

So I started a little POC regarding the service: 
I've just installed nodejs express and make Josh's asconfigc available via
an http endpoint.
This is just a few lines of code and it works as expected [1][2].
In the real world, the service would expect some FlexJS code and response
with the compiled code or an URL where the code lives.
Next step would be to build a FlexJS app that implements a CodePen inspired
UI and consumes the service. 

Before starting with this I'd like to ask if you think that this approach is
worth following it or if you expect too many issues (maybe also legal
issues) or if you think it's completely useless.

Of course, it's a long way to making it happen and a powerful backend is
needed and this is probably not for free. I'm not yet familiar enough with
the various cloud services out there but I could imagine that something like
'AWS lamda' would work. Maybe Apache Infra provide something like this?

Thanks,
Olaf

[1] https://snag.gy/Bb83sq.jpg
[2] https://snag.gy/crM43o.jpg



--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-POC-JSFiddle-for-FlexJS-Compile-as-a-service-tp61369.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.


Re: [FlexJS] Issue with JSON.stringify and Bindable VO objects

2017-04-26 Thread piotrz
Greg,

Ohhh! I know what you mean - bloody deadlines! :) Thank you!

Piotr



-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Issue-with-JSON-stringify-and-Bindable-VO-objects-tp61195p61368.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.


Re: [FlexJS] Two-way binding

2017-04-26 Thread piotrz
It would be good to have two-way binding working with -> @ - I've just raised
a jira [1]

[1] https://issues.apache.org/jira/browse/FLEX-35298

Piotr



-
Apache Flex PMC
piotrzarzyck...@gmail.com
--
View this message in context: 
http://apache-flex-development.247.n4.nabble.com/FlexJS-Two-way-binding-tp61314p61367.html
Sent from the Apache Flex Development mailing list archive at Nabble.com.