Wouldn't that mean that I have somehow on the client, and based on the browsers mime capabilities/agent type select the correct Sound object to play? That would seem to me to go against the GWT's core philosophy of writing code that is capable to run in all browsers, without having to go through such case by case specialized code.
And if I understood correct the multiple sources in HTML5, the selection should be done automatically and should be hidden from the programmer.
I also tried to use the addSound() method of the GWT's Audio class that you use for the HTML5 sound implementation in your library. But that didn't work either, although a simple HTML5 page with multiple sources for a given sound worked just fine on the same browser. Do you have any insight if there is such a known bug/problem with the Audio class? I clearly understand that gwt-voices tries to create a common denominator environment for the various technologies that exist to provide audio in a web app, but I was hoping that I could manipulate it to add some needed methods, if only for the HTML5 case, which could be deactivated for the rest.
Thanks for your answers and your time!
On Monday, March 5, 2012 11:05:17 PM UTC+2, Fred Sauer wrote:
Just create multiple Sound objects via SoundController#createSound
Is it possible somehow, through gwt-voices to add multiple sources for the same sound, like in HTML5 someone can has multiple source attributes to define different audio formats for the same
audio tag?
If not, is there a plan to add such a support?
I have seen that the Audio object, that gwt-voices is using internally for the Html5Sound case, does has such a method, i.e. addSource. Any reason why this
can't become visible in gwt-voices API as well?
Thanks in advance,
G.