Hi Michael. I've been bugging Brian over on the Blog about this here ... http://blogs.adobe.com/osmf/2009/08/osmf-v04-now-available.html
I'm going to be attempting to hack the OvpCCManager and OvpCuePointManager to allow us to use the old captioning classes in OSMF. Have you had any luck with this yourself? If so maybe we should put our heads together. Get in touch if you are interested.
[leeprobert AT gmail DOT com]
We haven't spec'd out exactly how cue points will work in OSMF, and how our support might differ from the OVP approach, but I suspect we'll build upon our core metadata support. The idea is that the cue points would be represented as temporal metadata, for which clients can listen to events and respond accordingly. The OVP approach is (IMO) elegant in that they define a base cue point manager which is essentially a temporal metadata manager, and which is flexible enough that they were able to build captioning support on top of that base manager class. One difference we might introduce with an OSMF-based solution is that we'd be adding/changing the metadata on a MediaElement, whereas OVP funnels it through the NetStream.
I haven't tried to hack it but I am prepared to give it a go so let me
know which approach you are taking
Micheal X Cronin
Thanks for your reply. I agree that AS cue points don't necessarily
need to work as they do in OVP but I do like their flexible approach.
I look forward to seeing your solution.
Hi guys. I will be attempting a hack today. My approach will be as follows ...
- Extend NetLoader (OSMF) and override methods for setting the NetStream object. Replace the NetStream object with the OvpNetStream Object.
- Provide an interface in the extended NetLoader object to pass the OvpNetStream object into the OvpCCManager class.
- Manage the CC as you would within the OVP framework.
Brian, without having to dig into the code too much, does this approach seem feasible?
By the time you come online later I might have this on my blog ...
Whoops ... my bad. Apologies.