Are there any similar implementation that you may advice me to check as
sample ?
14.02.2019 22:36 tarihinde Alex Harui yazdı:
So the progressHandler in URLLoader is dispatching a ValueEvent, not
the HTTPStatusEvent you want. The emulation HTTPService should
override what gets dispatched. You could override dispatchEvent and
dispatch the right thing.
HTH,
-Alex
*From: *Serkan Taş <serkan....@likyateknoloji.com>
*Reply-To: *"users@royale.apache.org" <users@royale.apache.org>
*Date: *Thursday, February 14, 2019 at 11:30 AM
*To: *"users@royale.apache.org" <users@royale.apache.org>
*Subject: *Re: Work on Emulation
Found it :
cid:part1.364CE0CC.0B2C681F@likyateknoloji.com
14.02.2019 22:28 tarihinde Serkan Taş yazdı:
Alex,
I am not sure that I know how to capture call stack in Firefox Dev
Edition.
Thanks,
Serkan
14.02.2019 22:23 tarihinde Alex Harui yazdı:
In Royale in the browser, some events aren’t what you think
they are. If you listen for a
org.apache.royale.events.MouseEvent, the browser dispatches a
browser MouseEvent and Royale runs code to catch that and
convert it to org.apache.royale.events.MouseEvent.
I suspect that the browser is sending its low-level event and
Royale code needs to be added to convert it to a Royale
HTTPStatusEvent. That’s why I suggested checking the call
stack as it might be possible to copy the MouseEventConverter
patterns to do the conversion.
HTH,
-Alex
*From: *Serkan Taş <serkan....@likyateknoloji.com>
<mailto:serkan....@likyateknoloji.com>
*Reply-To: *"users@royale.apache.org"
<mailto:users@royale.apache.org> <users@royale.apache.org>
<mailto:users@royale.apache.org>
*Date: *Thursday, February 14, 2019 at 11:14 AM
*To: *"users@royale.apache.org"
<mailto:users@royale.apache.org> <users@royale.apache.org>
<mailto:users@royale.apache.org>
*Subject: *Re: Work on Emulation
Hi Alex,
13.02.2019 01:50 tarihinde Alex Harui yazdı:
Hi Serkan,
In the emulation components, you may have to catch certain
events and redispatch them with the appropriate type. In
the event handler that calls Alert, what is the type of
the event parameter. Is it an HTTPStatusEvent or a
browser event?
Here is the registration :
loader.addEventListener(HTTPStatusEvent.HTTP_STATUS, httpResult);
It is registered as HTTPStatusEvent. And the handler has the
signature :
private function httpResult(event:HTTPStatusEvent):void
If it is an HTTPStatusEvent, then the emulation may be as
simple as adding a “status” getter that returns the
event’s value property.
As it is HTTPStatusEvent, than it will be enough to implement
your advice to the JS section of the code. But when I check
the code, there is getter for status :
COMPILE::JS
public class HTTPStatusEvent extends
org.apache.royale.events.Event
{
/* include "../core/Version.as"; */
public static const HTTP_STATUS:String = "httpStatus";
private var m_status:int;
//--------------------------------------------------------------------------
//
// Class constants
//
//--------------------------------------------------------------------------
//--------------------------------------------------------------------------
//
// Constructor
//
//--------------------------------------------------------------------------
/**
* Constructor.
*
* @param type The event type; indicates the action
that caused the event.
*
* @param bubbles Specifies whether the event can bubble
* up the display list hierarchy.
*
* @param cancelable Specifies whether the behavior
* associated with the event can be prevented.
*
* @langversion 3.0
* @playerversion Flash 9
* @playerversion AIR 1.1
* @productversion Royale 0.9.3
*/
public function HTTPStatusEvent(type:String,
bubbles:Boolean = false,
cancelable:Boolean = false, status:int = 0)
{
super(type, bubbles, cancelable);
m_status = status;
}
public function get status() : int
{
return m_status;
}
}
But returns internal m_status, which is in my case is undefined.
Thanks,
Serkan
If it is a browser event, then look at the call stack to
see who is dispatching the event. The code that
dispatches the event might need to create an
HTTPStatusEvent, copy the browser event properties into
it, and then dispatch the HTTPStatusEvent. If you are not
sure, post the call stack.
HTH,
-Alex
*From: *Serkan Taş <serkan....@likyateknoloji.com>
<mailto:serkan....@likyateknoloji.com>
*Reply-To: *"users@royale.apache.org"
<mailto:users@royale.apache.org> <users@royale.apache.org>
<mailto:users@royale.apache.org>
*Date: *Tuesday, February 12, 2019 at 11:52 AM
*To: *"users@royale.apache.org"
<mailto:users@royale.apache.org> <users@royale.apache.org>
<mailto:users@royale.apache.org>
*Subject: *Re: Work on Emulation
Hi Alex (only you are interested in emulation, right ?) :)))
While debugging my code, I realized that there is a
conflict between in status and httpStatus values of the
event HTTPStatusEvent, at least in my mind.
My old flex code, ported to royale, registers and event
listener for HTTPStatusEvent.HTTP_STATUS and handler
httpResult as below (loader is initialized somewhere) :
loader.addEventListener(HTTPStatusEvent.HTTP_STATUS,
httpResult);
After making http request, the handler returns as below on
screen :
cid:part1.78AD6179.FDDC1C04@likyateknoloji.com
This is where it fires window :
cid:part2.FC1F7F03.0A76E9BA@likyateknoloji.com
It is normally expected that the status holds real value,
but here it is not, httpStatus contains the value in
"value" as below :
cid:part3.E6732EF2.3364FCC2@likyateknoloji.com
As the HTTPStatusEvent is emulated, I am not sure how to
go to solve the issue.
Thanks,
Serkan
