- Install failure [1 Update]
- reflection [3 Updates]
- emulator not showing app [1 Update]
- Toast not disappearing [2 Updates]
- Clear calling activity on finish [1 Update]
- Gingerbread BroadcastReceiver Issue [1 Update]
- WebView document.implementation.hasFeature("XPath", "3.0") should not return true. [1 Update]
- User can purchase other apps but not mine? [1 Update]
- Renderscript documentation [1 Update]
- ProgressDialog Theme - TextColor or Background but not both [1 Update]
- Installpackged [1 Update]
- Can I read .pdf From assert folder? [1 Update]
- kypriakos <firstname.lastname@example.org> Feb 28 10:09PM -0800 ^
I have two separate projects each compiling and executing fine on
Android devices (currently this code calls for Android 1.6 so I am
running on that). I needed to reuse some of the classes from one
project into another so I brought over their source code and
recompiled. Things compile fine but during the installation of the apk
I get the [INSTALL_FAILURE_DEXOPT]. This means that either there's no
space left to install this on the device (which is not the case here)
or the verification failed. What would cause a verification error in
this case? I see below that the Package has a mismatched uid on disk
vs in settings? What does that designate?
Thanks in advance
D/AndroidRuntime( 499): >>>>>>>>>>>>>> AndroidRuntime START
D/AndroidRuntime( 499): CheckJNI is OFF
D/AndroidRuntime( 499): --- registering native functions ---
I/jdwp ( 499): received file descriptor 20 from ADB
D/ddm-heap( 499): Got feature list request
D/PackageParser( 85): Scanning package: /data/app/vmdl13009.tmp
D/dalvikvm( 85): GC freed 19996 objects / 964408 bytes in 247ms
E/PackageManager( 85): Package com.WebCP_16 has mismatched uid:
10044 on disk, 10047 in settings
I/PackageManager( 85): /data/app/vmdl13009.tmp changed; unpacking
D/installd( 53): DexInv: --- BEGIN '/data/app/vmdl13009.tmp' ---
W/dalvikvm( 506): partial write in inflate (24536 vs 32768)
W/dalvikvm( 506): DexOptZ: extraction of classes.dex from /data/app/
W/installd( 53): DexInv: --- END '/data/app/vmdl13009.tmp' ---
status=0xff00, process failed
E/installd( 53): dexopt failed on '/data/dalvik-cache/
data@email@example.com@classes.dex' res = 65280
W/PackageManager( 85): Package couldn't be installed in /data/app/
D/AndroidRuntime( 499): Shutting down VM
D/dalvikvm( 499): DestroyJavaVM waiting for non-daemon threads to exit
- Zsolt Vasvari <firstname.lastname@example.org> Feb 28 08:47PM -0800 ^
The answer is, of course, not. They are probably using different
The better question is, why would you want to do this?
- Hari Edo <email@example.com> Feb 28 08:54PM -0800 ^
Reflection is the act of asking the JVM (Dalvik VM) and the
current ClassLoader object instance for information about the
classes it can reach in memory or along the Java CLASSPATH.
It has nothing to do with inter-process communication. If it
is not in the current process, or reachable by the CLASSPATH,
it won't be found.
On Android, different package names are given different user
ids, and apps reside in completely different directories, to
further limit the visibility of files related to a different
- Krt <firstname.lastname@example.org> Feb 28 09:38PM -0800 ^
Thanks for the response.
I am trying to alter something in android framework source , and this
calls for an addition of a new class in framework source.
But I did not want to add any new source under framework, instead
planned to have it under vendor and load that new class dynamically in
So I was thinking if reflection will come in handy.
- "sourabh '''72'''" <email@example.com> Feb 28 09:45PM -0800 ^
can you please more elaborate on " launcher thing " like how to open
and how to play with that
and please give the procedure for fixing my problem
- Kevin TeslaCoil Software <firstname.lastname@example.org> Feb 28 09:01PM -0800 ^
I think I've seen something like if you show a toast and then your
activity is destroyed before the toast disappears it might get stuck.
Easier to reproduce if you're calling finish(). Probably could avoid
it if you save the toast reference and do a .cancel() in onDestroy.
- Zsolt Vasvari <email@example.com> Feb 28 09:22PM -0800 ^
On Mar 1, 1:01 pm, Kevin TeslaCoil Software <ke...@teslacoilsw.com>
> activity is destroyed before the toast disappears it might get stuck.
> Easier to reproduce if you're calling finish(). Probably could avoid
> it if you save the toast reference and do a .cancel() in onDestroy.
That's absolutely not the case. A lot of my activities finish with a
toast and I've never seen one stuck.
- Manish Garg <firstname.lastname@example.org> Feb 28 09:10PM -0800 ^
- Dianne Hackborn <email@example.com> Feb 28 09:09PM -0800 ^
It's supposed to be called (and if there isn't an actual pending command to
deliver it is called with a null Intent), but there is a bug on 2.3 where it
is not. It will only be called in cases where there are actual outstanding
intents to deliver.
Android framework engineer
Note: please don't send private questions to me, as I don't have time to
provide private support, and so won't reply to such e-mails. All such
questions should be posted on public forums, where I and others can see and
- Heath Borders <firstname.lastname@example.org> Feb 28 08:53PM -0800 ^
However, the following are undefined:
I'm running API Level 7 in the emulator.
- Zsolt Vasvari <email@example.com> Feb 28 08:45PM -0800 ^
> I wouldn't recommend that... I have had a couple people actually update
> their information and then the sale goes through.
By the time I see it, it's usually hours after the attempted
purchase. And they can always repurchase (and usually do).
- Peter Eastman <firstname.lastname@example.org> Feb 28 08:43PM -0800 ^
Now that the 3.0 SDK is out, I've read through the Renderscript
documentation. But it's pretty sketchy, and still doesn't answer most
of my questions.
Graphical scripts seem to consist of shader definitions and drawing
code. But I assume the shaders ultimately get compiled down to the
same machine code whether you define them with Renderscript or GLSL,
and the drawing code presumably gets compiled down to roughly the same
set of OpenGL calls that you'd make in Java or C++. So what are the
advantages of using Renderscript instead?
Then there are compute scripts, which I gather are meant to be a GPGPU
framework? I have a lot of experience with CUDA and OpenCL, so I'd
like to learn how it compares with those. There's hardly any
documentation on writing compute scripts, and only a single trivial
example. What threading model does it use? Does it support inter-
thread communication? Scattered writes?
- William Ferguson <email@example.com> Feb 28 05:35PM -0800 ^
> style="@android:style/Widget.ProgressBar"*, as was shown in that
> stackoverflow post? You posted samples show AlertDialog and plaing Dialog.
> No sure if it matters, but worth a shot if you didn't try it already.
I'm pretty certain I covered *exactly* that in one of the many
hundreds of scenarios I ran over the weekend.
But I'll go back and give it one more shot.
> This is not something I've had to do, but a custom layout and setting it
> with setView() on the dialog seems like the simplest alternative.
This is where I'm at too.
It just doesn't feel right to ditch exactly the layout I want and
replace it with the same layout but with a different style. All
because I can't seem to get my style applied correctly.
- TreKing <firstname.lastname@example.org> Feb 28 02:43PM -0600 ^
> we as a manufacture have permissions to use methods which are not available
> to end-client users/developers.
You should try a different group then - try platform or porting. This is for
using the SDK to do "official" things.
TreKing <http://sites.google.com/site/rezmobileapps/treking> - Chicago
transit tracking app for Android-powered devices
- Sunil Lakhiyani <email@example.com> Feb 28 12:29PM -0800 ^
I have .pdf in assert folder, also i have pdf viewer ad acrobat reader
installed in my phone.
Can i read .pdf from assert anyway (using those 2)?
You received this message because you are subscribed to the Google
Groups "Android Developers" group.
To post to this group, send email to firstname.lastname@example.org
To unsubscribe from this group, send email to
For more options, visit this group at