Re: [Laszlo-user] Problem with rtmpconnection (to red5) and 4.0.12

[André Bargull]

Created LPP-5930 ("http://www.openlaszlo.org/jira/browse/LPP-5930";) to 
track this issue.

Hmm, we rewrote how the base "schema" is declared to the compiler,and that
does indeed declare "call" as a final method:
<interface name="Instance" extends="Object">
  <method name="validateClassStructure" final="true"/>
  <method name="initialize" />
  <method name="nextMethod" final="true"/>
  <method name="addProperty" final="true"/>
  <method name="constructor" final="true"/>
  <method name="hasOwnProperty" final="true"/>
  <method name="isPropertyEnumerable" final="true"/>
  <method name="isPrototypeOf" final="true"/>
  <method name="toLocaleString" final="true"/>
  <method name="prototype" final="true"/>
  <method name="apply" final="true"/>
  <method name="call" final="true"/>
</interface>

Since 'call' is a javascript built-in, it would really be best to rename
this method in the library.




On Fri, May 2, 2008 at 1:11 PM, Anthony Bargnesi <abargnesi at gmail.com 
<http://www.openlaszlo.org/mailman/listinfo/laszlo-user>>
wrote:

>/ I found out some further information.  I modified
/>/ components/extensions/av/rtmpconnection.lzx to make the netconnection always
/>/ through the line:
/>/
/>/ this._nc.call(func, obj);
/>/
/>/ instead of:
/>/
/>/ this._nc.call.apply(arguments);
/>/
/>/ This condition takes place at line 233 of rtmpconnect.lzx.  It seems like
/>/ the compiler warnings say that the "call" method cannot be overriden in a
/>/ base class.  So by reverting back to "this._nc.call(func, obj);" does the
/>/ trick, but obviously this refactor was done for a reason.  Plus I still get
/>/ the compiler warnings.
/>/
/>/ Hope this helps uncover some issues,
/>/ Tony
/>/
/>/
/>/ On Fri, May 2, 2008 at 8:26 AM, Anthony Bargnesi <abargnesi at gmail.com 
<http://www.openlaszlo.org/mailman/listinfo/laszlo-user>>
/>/ wrote:
/>/
/>/ > Hello,
/>/ >
/>/ > I recently ran a lzx test app in 4.0.12 that makes a call to Red5.  What
/>/ > worked fine in an previous laszlo versions
/>/ > has since stopped working.  This time, however, I am seeing the
/>/ > following compiler warnings:
/>/ >
/>/ > *extensions/av/rtmpconnection.lzx:14:49:* Method rtmpconnection.call is
/>/ > overriding a superclass method of the same name which has been declared
/>/ > non-overridable
/>/ > *extensions/av/rtmpconnection.lzx:250:48:* Method netremotecall.call is
/>/ > overriding a superclass method of the same name which has been declared
/>/ > non-overridable
/>/ >
/>/ > Here is the example LZX:
/>/ >
/>/ > <canvas debug="true">
/>/ >     <button name="callButton" onclick="onCardEventCall();">
/>/ >         Call sendEvent
/>/ >         <method name="onCardEventCall">
/>/ >             canvas.rtmp.onCardEvent.event = "button was clicked!";
/>/ >             canvas.rtmp.onCardEvent.call();
/>/ >         </method>
/>/ >     </button>
/>/ >     <rtmpconnection name="rtmp" debug="true" autoconnect="true"
/>/ > src="rtmpt://localhost:8888/srvr/myhome" 
<rtmpt://localhost:8888/srvr/myhome%22>>
/>/ >         <handler name="onconnect">
/>/ >             Debug.write("connected");
/>/ >         </handler>
/>/ >         <handler name="onerror">
/>/ >             Debug.write("error ",this.status);
/>/ >         </handler>
/>/ >
/>/ >        <netremotecall name="onCardEvent" funcname="onCardEvent">
/>/ >             <attribute name="event" value="" type="string" />
/>/ >             <netparam name="vars1"><method name="getValue">return
/>/ > parent.event;</method></netparam>
/>/ >         </netremotecall>
/>/ >
/>/ >        <netremotecall name="newEvent" funcname="newEvent">
/>/ >             <method name="onResult" args="value">
/>/ >                 Debug.write('event received: ' + value);
/>/ >             </method>
/>/ >         </netremotecall>
/>/ >     </rtmpconnection>
/>/ > </canvas>
/>/ >
/>/ > I see the rtmpconnection being made on the Red5 server, but when I try
/>/ > to send an event by clicking the "callButton"
/>/ > it never comes through in Red5.  I'm thinking this has something to do
/>/ > with the compiler warnings.
/>/ >
/>/ > Any thoughts?
/>/ >
/>/ > Tony
/>/ >
/>/
/>/
/

--
Henry Minsky
Software Architect
hminsky at laszlosystems.com 
<http://www.openlaszlo.org/mailman/listinfo/laszlo-user>