I would try to leverage flexbox for layout and stay away from absolute values.
Things need to be fluid and responsive to support the many devices.
Richard Oren
714-916-2102
> On 7 Aug 2015, at 9:50 pm, Alex Harui wrote:
>
> One of my goals for FlexJS is to leverage as much code in the browser
> That’s convenient, but my main point was to reduce the number of
> databinding expressions in MXML that have to be written by the developer,
> and set up and run at startup.
Oh yeah, you are right, I didn't it read well, but still, usually for a
library, even more a public one, I can pay a bit
On 8/11/15, 8:15 AM, "Frédéric THOMAS" wrote:
>>Also, another point I just remembered is MXML. I don’t know that
>> MouseButton would actually show up in MXML attribute values but for
>>other
>> things it might be better to use simple String constants and do extra
>> checking inside so your MXM
> I think the definition of consistent and safe might have some trade-offs.
> The resulting production code still needs to be small and fast so I tend
> to prefer solutions where you can remove development-time checking you
> don’t need I production.
I understand your point.
> Also, another point
On 8/11/15, 6:37 AM, "Frédéric THOMAS" wrote:
>> but no objection if that’s what folks want.
>
>My first reaction when it comes to a public, is to offer a consistent,
>safe and clear API but I'm open to anything too.
I think the definition of consistent and safe might have some trade-offs.
The
> but no objection if that’s what folks want.
My first reaction when it comes to a public, is to offer a consistent, safe and
clear API but I'm open to anything too.
Btw, I noticed the package, class and Object.defineProperties comments are not
copied to the generated JS, so neither the Apache
On 8/11/15, 6:06 AM, "Frédéric THOMAS" wrote:
>> Makes sense. I was just wondering why a MouseButton class is needed at
>> all and not just a plain String constant.
>
>Maybe is not, but I added, maybe I shouldn't had, isButton(button) =>
>boolean which tests to see which button was pressed duri
> Makes sense. I was just wondering why a MouseButton class is needed at
> all and not just a plain String constant.
Maybe is not, but I added, maybe I shouldn't had, isButton(button) => boolean
which tests to see which button was pressed during the event, the type of
button is goog.events.Brows
On 8/11/15, 5:31 AM, "Frédéric THOMAS" wrote:
>Btw, I can now produce the Core.swc filled with the compiled AS3 classes
>and the JS in a once and in IntelliJ, indeed, I only migrated 1 JS file
>to AS but it is what I wanted first, validate I can do it, so, I will
>continue on it.
Awesome!
-A
On 8/11/15, 5:22 AM, "Frédéric THOMAS" wrote:
>> What is the advantage of doing it this way?
>
>My point in migrating the framework is also to figure out how a framework
>developer would feel comfortable working with, using and developing new
>component seeing how the others are developed, for
> No idea. You did remind me that I never finished investigating whether
> there was a valid AS3 syntax for disambiguating a class without a package
> and a class with a package. Maybe the best thing for you right now is to
> not worry about the conflict and see if there are other issues? Or does
>
> Well, try running ‘ant main’, then in examples/DataBindingTest run Ant
> again and see if you get any errors or warnings. Maybe something changed
> so that isn’t needed anymore. You might just want to use a new
> compile-js-config.xml file.
Until Core is not totally migrated, that won't be possi
12 matches
Mail list logo
