Fixes JNI Bindings & startup perf.

Fixes JNI Bindings & startup perf.

This patch follows digit's suggestions:
- Removes ScopedJavaObject for jclasses.
- Uses const char kFoo[] rather than pointers.

BUG=
TEST=base/android/jni_generator/jni_generator_tests.py


Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=124413

[bulach, 2012-02-28 10:59:50]

mark: a few updates for the jni generator.. fyi, joth and pliard have reviewed
downstream and it has been cycling for a few days in our trees..

btw, I hope you won't get bored with us ;) but if you do, please let me know and
I'll distribute the load to other base/ OWNERS.

thanks!

[joth, 2012-02-28 11:37:25]

lgtm

[Philippe, 2012-02-28 12:01:46]

On 2012/02/28 11:37:25, joth wrote:
> lgtm

LGTM.

[Mark Mentovai, 2012-02-28 14:44:29]

http://codereview.chromium.org/9466024/diff/1/base/android/jni_generator/gold...
File base/android/jni_generator/golden_sample_for_tests_jni.h (right):

http://codereview.chromium.org/9466024/diff/1/base/android/jni_generator/gold...
base/android/jni_generator/golden_sample_for_tests_jni.h:25: static const char
kInnerClassClassPath[] =
Drop the “static,” which has no effect in this case because you’re in an
anonymous namespace.

I know this is a generated file, so only make this change in the generator if it
will always put this inside an anonymous namespace.

http://codereview.chromium.org/9466024/diff/1/base/android/jni_generator/gold...
base/android/jni_generator/golden_sample_for_tests_jni.h:27: static const char
kSampleForTestsClassPath[] =
Same here.

http://codereview.chromium.org/9466024/diff/1/base/android/jni_generator/gold...
base/android/jni_generator/golden_sample_for_tests_jni.h:88: /* Must call
RegisterNativesImpl()  */
Would it be helpful to have these make any “register” calls that haven’t been
made yet themselves instead of imposing the requirement on a caller?

[Mark Mentovai, 2012-02-28 14:44:50]

(LGTM with the two “static” changes, if needed.)

[bulach, 2012-02-28 15:11:36]

thanks again for the super quick review mark! 
"static" removed and one clarification below, please let me know if it makes
sense..

http://codereview.chromium.org/9466024/diff/1/base/android/jni_generator/gold...
File base/android/jni_generator/golden_sample_for_tests_jni.h (right):

http://codereview.chromium.org/9466024/diff/1/base/android/jni_generator/gold...
base/android/jni_generator/golden_sample_for_tests_jni.h:25: static const char
kInnerClassClassPath[] =
On 2012/02/28 14:44:30, Mark Mentovai wrote:
> Drop the “static,” which has no effect in this case because you’re in an
> anonymous namespace.
> 
> I know this is a generated file, so only make this change in the generator if
it
> will always put this inside an anonymous namespace.

done. yes, it always generates in an anon namespace..

http://codereview.chromium.org/9466024/diff/1/base/android/jni_generator/gold...
base/android/jni_generator/golden_sample_for_tests_jni.h:27: static const char
kSampleForTestsClassPath[] =
On 2012/02/28 14:44:30, Mark Mentovai wrote:
> Same here.

Done.

http://codereview.chromium.org/9466024/diff/1/base/android/jni_generator/gold...
base/android/jni_generator/golden_sample_for_tests_jni.h:88: /* Must call
RegisterNativesImpl()  */
On 2012/02/28 14:44:30, Mark Mentovai wrote:
> Would it be helpful to have these make any “register” calls that haven’t
been
> made yet themselves instead of imposing the requirement on a caller?

hmm.. this "Register" calls are required at startup for a variety of reasons...
we haven't got a clear / automatic registration mechanism just yet, but it'd be
an error to not (manually) call it..

so this message is just so developers could learn what they need to do should
they hit this check.. 

we also didn't want to add as a message to avoid adding more strings, as again,
it should be impossible to hit such error except during early development..

does it make sense?

[Mark Mentovai, 2012-02-28 15:14:12]

LGTM

[commit-bot: I haz the power, 2012-03-01 13:48:28]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/bulach@chromium.org/9466024/4003

[commit-bot: I haz the power, 2012-03-01 15:51:52]

Change committed as 124413