Manifest file

by senrooy » Thu, 17 Sep 2009 19:01:09 GMT

Sponsored Links

I am new to android. I am using SDK and eclipse with ADT plug-ins for
the first time. I want to know what is the significance of the
AndroidManifest.xml file which is created automatically when we create
an android project. Can we modify this file for any reason?

Thanks n regards,

Manifest file

by Ajaz Ahmed » Thu, 17 Sep 2009 19:07:34 GMT


Yes we can modify, but i think its available to modify.
we can add activities references, user permissions,event receivers.
AndroidMasifest.xml file is same like struts.xml file in struts2.

Imtiaz Ali Shah


Sponsored Links

Manifest file

by Mark Murphy » Thu, 17 Sep 2009 19:07:53 GMT


There is a Web page in the Android documentation that covers this: 

You can modify it when you need to. If you are using Eclipse, it may
make some modifications for you, but other changes you may need to make

Mark Murphy (a Commons Guy)  | 

_Android Programming Tutorials_ Version 1.0 In Print!


Manifest file

by senrooy » Thu, 17 Sep 2009 19:22:11 GMT

 In which cases I can modify this file? I have done some basic android
applications but I didn't modify this file and applications worked
fine. For what reason we should modify this file?


Manifest file

by Naveen Krishna Ch » Thu, 17 Sep 2009 19:25:47 GMT

 For the first time when we develop an activity, The eclipse automatically
adds it to
the AndroidManifest.xml. As we want to add more activities or any frame work

We need to add the, we can add permissions, intent filter, etc. also

2009/9/17 senrooy <>

Shine bright,
(: Naveen Krishna Ch :)


Other Threads

1. SoundPool resume problem (after playing a second sound while looped sound is paused)

I've been using SoundPool for awhile now with no issues to speak of,
but I recently added my first 'looped' sound, which led to the use of

My problem is that while this works

streamId = play(...)  // loop = -1
pause( streamId )
resume( streamId )


This seems to fail

streamId = play()
pause( streamId )
resume( streamId )

does nothing.


In the log I generally see some garbage collection occuring in the
second case which may not be happening in the first case.  It
SEEEEEEEEMs like playing the second sound, while the first is paused,
somehow damages the first sound's stream.  Trying to restart the first
sound with a 'play' also fails (in general, trying to 'play' a
'paused' sound appears to fail, requiring a 'resume' instead)

And 'stopping' the sound works to unjam it, but, of course, forces you
to reload the samples.

I've added enough logging that I am fairly sure I am not crossing my
streamIds, as it were.

Oh, and only the first sound is looped.  All other sounds are one-

At first I thought it might have to do with the length of the looped
sound, but I have the same problem if I use a small sound for the
loop.  All sounds are 22KHz, mono, 16bit, from OGG resources.


And yes, I *did* chase my tail for awhlie thanks to some initial
confusion over sound Ids and stream Ids. (I think a few words in the
ref documents could be improved :-)

Actually, I wish there were a 'stop' that didn't release the samples,
as that's what I really want to do (not pause/resume)... some form of
pause/rewind/resume at the very least.

I've tried various forms of setLoop, thinking that perhaps it is just
losing that state along the way, but no joy.

Thanks in advance.


2. RIL - GSM Cell broadcast related query

Hi all,

I have some doubt on how to package the response value for the GSM
cell broadcast configuration use case.

According to 3GPP TS 27.007, the AT cmd to Select Cell Broadcast
Message Types +CSCB, is of the following format
for eg to get the CBS configuration
+CSCB: <mode>,<mids>,<dcss>
eg: +CSCB: 0,"0,1,5,320-478", "0-3,5"

If we take the <mids> in the above example, the service id values can
be of both range values (320-478) and specific service id
Same goes for the <dcss> code scheme values.

"data" is a const RIL_GSM_BroadcastSmsConfigInfo **
"datalen" is count * sizeof(RIL_GSM_BroadcastSmsConfigInfo *)

The struct RIL_GSM_BroadcastSmsConfigInfo has following parameters

typedef struct {
    int fromServiceId;
    int toServiceId;
    int fromCodeScheme;
    int toCodeScheme;
    unsigned char selected;
} RIL_GSM_BroadcastSmsConfigInfo;

In the above structure it accepts only from and to range values. This
leaves only the following way to pack the values and send.

count = 4 (mids count = 4, dcss count =2)

fromServiceId: 0 toServiceId: 0
fromCodeScheme: 0 toCodeScheme: 3

fromServiceId: 1 toServiceId: 1
fromCodeScheme: 5 toCodeScheme: 5

fromServiceId: 5 toServiceId: 5
fromCodeScheme: 0 toCodeScheme: 0

fromServiceId: 320 toServiceId: 478
fromCodeScheme: 0 toCodeScheme: 0

This may not be the right way, as it is difficult for the client app
to interpret it.

Shouldn't be easier to have represent both service id and code scheme
values as strings in the RIL_GSM_BroadcastSmsConfigInfo pass the
<mids> and <dcss> values as such, rather than from to values
for eg:
typedef struct {
    char* serviceId;
    char*  codeScheme;
    unsigned char selected;
} RIL_GSM_BroadcastSmsConfigInfo;

or to represent it as integer arrays and fill in the array with
corresponding values(convert range values to specific values, for eg:
1-5 -> 1,2,3,4,5) and include both the service id and code scheme
counts as part of   RIL_GSM_BroadcastSmsConfigInfo ?

for eg:
typedef struct {
    int* serviceId;
    int* codeScheme;
    int serviceIdCount;
    int codeSchemeCount;
    unsigned char selected;
} RIL_GSM_BroadcastSmsConfigInfo;



3. Voice Mail Notifications

4. Underclocking to simulate other phones

5. download Attachment from email.

6. CheckBox spacing

7. What's wrong with this maps intent?