As Ravi mentioned the OpenCore H324M stack is very close to be released. It is used as part of the PV 2way engine which handles the management of the multimedia nodes similar to the way the player and author engines handle those use-cases in OpenCore today. The graph (i.e., the collection of multimedia nodes and their connections) used by the 2way engine is more complicated than either the author or player because it is handling both decoding and encoding tasks with real-time bidirectional communication. If you are only interested in playback use-cases, then the 2way engine is probably more complicated than you need and may not be the best reference. For the playback use-cases, a customized source node may be the best approach as freepine mentions. The source node would allow you to encapsulate the method of obtaining the data as well as the method of demuxing the data into one of the supported audio and video formats. If you only need to encapsulate the transport and the data will be in an already supported file format, another option is to look at the data stream interface in pvmi_data_stream_interface.h.
On Jan 8, 10:54 pm, Girish <[email protected]> wrote: > Hi Ravi Dave freepine, > > Thanks for your replies. > > If you can tell us the rough time line it will be of great help for > us. > > Also please let me know how OpenCORE H324M interacts with PV > multimedia framework. Is H324M stack created as a source node and fed > to decoder and finally rendered ? > > How exactly this is taken care ? Are there any licensing and royalty > to be paid incase we use OpenCORE MMFW, codecs and OpenCORE VT for > commercial product ? > > Do we need to publish the source code to android community if > integrate proprietary stack code and make a product out of it ? > > Please clarify. > > Regards > Girish > > On Jan 8, 11:04 pm, rktb <[email protected]> wrote: > > > Very soon :-) > > > On Jan 9, 8:21 am, freepine <[email protected]> wrote: > > > > Thanks, Dave. that's great. > > > Is there a planed timeline for it? > > > > On Fri, Jan 9, 2009 at 11:16 AM, Dave Sparks > > > <[email protected]>wrote: > > > > > This sounds a little like a H.324M stack. If so, you should know that > > > > Packet Video is planning to supply an H.324M stack for OpenCore. > > > > > On Jan 7, 11:05 pm, freepine <[email protected]> wrote: > > > > > You mean your case is similar with the URLs of HTTP download or RTSP > > > > > playback in Opencore, right? > > > > > If so, I think you might need to implement a customized source node to > > > > > retrieve encoded data from modem and integrated into opencore > > > > > framework. > > > > You > > > > > can refer to the implementation of PVMFDownloadManagerNode in > > > > > external/opencore/nodes/pvdownloadmanagernode. > > > > > > Then you can configure the node registry to create your customized > > > > > node > > > > when > > > > > recognized some special URL like "modem://xxx"... > > > > > > Just my 2 cents, > > > > > -freepine > > > > > > On Thu, Jan 8, 2009 at 12:43 PM, Girish <[email protected]> wrote: > > > > > > > Hi Dave and freepine, > > > > > > > In my case the encoded data to ¨my native c code¨ is received from > > > > > > modem. Which is nothing but ENCODED FRAME. > > > > > > > But in case if i pass this encoded frame data to the application, > > > > > > media player cant decode this data. As it either expects a URL or > > > > > > file > > > > > > stream. > > > > > > > This scenario nearly meets the requirement of passing URL to the > > > > > > media > > > > > > player. > > > > > > > In URL case also do modem gets the data from network sends the data > > > > > > from network to MMFW or to Application ? How this is taken care ? > > > > > > > Below is my understanding of the android architecture. > > > > > > > ---------------------------------------- > > > > > > Application > > > > > > ---------------------------------------- > > > > > > Application framework > > > > > > ---------------------------------------- > > > > > > native applications > > > > > > > ******************** **************** > > > > > > * my_native_c_code *====> * MMFW * > > > > > > ******************** **************** > > > > > > > ---------------------------------------- > > > > > > Kernel > > > > > > > ====================================================== MODEM- > > > > > > APPLICATION interface > > > > > > > MODEM > > > > > > > --------------------------------- > > > > > > > Please clarify if my understaning is wrong. > > > > > > > Regards > > > > > > Girish > > > > > > > On Jan 7, 9:15 pm, freepine <[email protected]> wrote: > > > > > > > I wrote a simple native app for testing only. With it PV can > > > > construct > > > > > > node > > > > > > > graph and start data flow correctly, but obviously it can't > > > > > > > display > > > > video > > > > > > > frames on UI:) You can dump decoded output to a file and analyze > > > > > > > it > > > > > > manually > > > > > > > or play it with YUV player. just FYI. > > > > > > > BTW, opencore itself brings with a unit test framework under > > > > > > > external/opencore/engines/player/test. You can build it by > > > > uncommenting > > > > > > its > > > > > > > make file in external/opencore/Android.mk. > > > > > > > > ======================================= > > > > > > > int main(int argc, char** argv) > > > > > > > { > > > > > > > sp<ProcessState> proc = ProcessState::self(); > > > > > > > proc->startThreadPool(); > > > > > > > MediaPlayer mediaplayer; > > > > > > > if(argc > 0) > > > > > > > { > > > > > > > LOGI("set datasource: %s", argv[0]); > > > > > > > mediaplayer.setDataSource(argv[1]); > > > > > > > } > > > > > > > else > > > > > > > { > > > > > > > LOGI("set default datasource: /data/test.mp4"); > > > > > > > mediaplayer.setDataSource("/data/test.mp4"); > > > > > > > } > > > > > > > sp<SurfaceComposerClient> client = new SurfaceComposerClient; > > > > > > > int pid = getpid(); > > > > > > > sp<Surface> surface(client->createSurface(pid, 0, 176, 144, > > > > > > > PIXEL_FORMAT_OPAQUE, > > > > > > > ISurfaceComposer::eFXSurfaceNormal|ISurfaceComposer::ePushBuffers)); > > > > > > > mediaplayer.setVideoSurface(surface); > > > > > > > mediaplayer.prepare(); > > > > > > > mediaplayer.start(); > > > > > > > for(int i=0; i<10; i++) > > > > > > > { > > > > > > > sleep(1); > > > > > > > LOGI("playing, %d seconds\n", i); > > > > > > > } > > > > > > > mediaplayer.stop(); > > > > > > > LOGI("quiting..."); > > > > > > > > } > > > > > > > > On Thu, Jan 8, 2009 at 10:02 AM, Dave Sparks < > > > > [email protected] > > > > > > >wrote: > > > > > > > > > That's not a scenario we plan to support. > > > > > > > > > At the very least, you need a thin Java app that interacts with > > > > > > > > all > > > > > > > > the API's that are not available in native code. If you want a > > > > > > > > pure > > > > > > > > native app, you are going to end up writing a lot of native > > > > > > > > shims > > > > to > > > > > > > > talk to Java code. > > > > > > > > > On Jan 7, 2:51 am, Girish <[email protected]> wrote: > > > > > > > > > Dear all, > > > > > > > > > > In my requirement i have to access OpenCORE MMFW from my > > > > > > > > > native c > > > > > > code > > > > > > > > > for decoding video data . Is it possible to access this way ? > > > > > > > > > > Is OpenCORE MMFW exposes any APIs for accessing these decoder > > > > APIs ? > > > > > > > > > like initdecoder, decode_one_frame , deinitdecoder ? which > > > > > > > > > file > > > > > > > > > exposes this kind of APIs. Please give some clues. > > > > > > > > > > Even i want to display the decoded data on to UI . Is this > > > > possible? > > > > > > > > > > Regards > > > > > > > > > Girish --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "android-framework" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/android-framework?hl=en -~----------~----~----~----~------~----~------~--~---
