Gtalk client implememtation

by junoon » Sun, 15 Mar 2009 15:52:35 GMT

Sponsored Links
 Can some tell me how to implement the gtalk client on android. i have
face a lot of problems implementing the unofficial source of the gtalk
client. The issues is related with xmpp classes.

Other Threads

1. How the video display size and orientation is set under Android?

I'm a newbie to Android and want to explore the video display flow
under Android.

I browsed the video surface output in opencore, and found that it can
not get the full video display info such as coordinate and
orientation. So it just allocate buffers from pmem heap according to
the original video size and post the buffer out of opencore ( to a
layer buffer module).

So my question is:
(1) Where the video display size and orientation?

(2) which module does the scaling and rotation? by hardware or

(3) Can anybody give a further explanation how the surface flinger and
layer buffer work?

(4) is there any way to let famework base libs (surface flinger and
layer buffer) output some log info like opencore?
     It's will be very helpful for me to trace the code.

Many thanks

2. Change video hardware on the fly

I need to change the video hardware in an android system on the fly
and I need the system, and any running applications, to recognize any
change in video size, format and buffers. I have been browsing the
source for the SurfaceFlinger, Clinet, Layer, DisplayHardware and
EGLDisplaySurface classes, but have been unable to figure out how to
do this. Can anyone give me a hint?

I created a second frame buffer driver in the kernel mapped to /dev/

I modified DisplayHardware and EGLDisplaySurface so that
EGLDisplaySurface uses a frame buffer based on the screen index that
SurfaceFlinger passes to DisplayHardware (i.e. if DisplayHardware is
passed 0 for the dpy parameter of its constructor then
EGLDisplaySurface will use fb0). I created a second GraphicPlane in
SurfaceFlinger using a second DisplayHardware object mapped to fb1
through EGLDisplaySurface .

I added a command to SurfaceFlinger's onTransact method to implement
the switch.

When the command to switch comes in I initialize the second display if
it has not been initialized before. Then I enumerate the existing
Layer objects calling each of their setSizeChanged and doTransaction
methods. I then call DisplayHardware's makeCurrent method for the new

Am I headed down the wrong path here or am I just missing a step (or
two, or ...)?

3. Where is SurfaceFlinger service is generated?

4. Banned from Android Community

5. comining/merging two images

6. Question on the BlogPost of ListView posted by Romanian Guy

7. Fake Character Device Driver