Change UMA proto product field to be an int32.

Change UMA proto product field to be an int32.

This makes it easier for downstream clients of the
component to add new product types without needing
to update the component source.

Also, adds a client method to return the product, so
that different MetricsService clients can provide
different values.

Also cleans up some unnecessary metrics:: namespace
prefixes.

BUG=415581
Committed: https://crrev.com/4c1d1ef9c811b58e83bff3e9cd8012d801bfa88b
Cr-Commit-Position: refs/heads/master@{#297254}

[Alexei Svitkine (slow), 2014-09-18 14:47:59]

Patchset #1 (id:1) has been deleted

[Alexei Svitkine (slow), 2014-09-25 16:42:39]

Patchset #1 (id:20001) has been deleted

[Alexei Svitkine (slow), 2014-09-25 16:55:31]

asvitkine@chromium.org changed reviewers:
+ gunsch@chromium.org, isherman@chromium.org

[gunsch, 2014-09-25 16:57:37]

On 2014/09/25 16:55:32, Alexei Svitkine wrote:

chromecast/ lgtm

[Ilya Sherman, 2014-09-25 18:29:55]

Is it worth adding a test to verify that the default product value is not set,
and that others are?  Otherwise, LGTM % nits, assuming the internal proto change
is approved.  Thanks, Alexei :)

https://codereview.chromium.org/573403002/diff/40001/components/metrics/metri...
File components/metrics/metrics_log.cc (right):

https://codereview.chromium.org/573403002/diff/40001/components/metrics/metri...
components/metrics/metrics_log.cc:130:
prefs::kStabilityIncompleteSessionEndCount, 0);
nit: Looks like this and several other lines below might not need to wrap
anymore.  However, I'd slightly prefer that you remove the "metrics::" namespace
references in a separate CL, to keep this one less visually cluttered.

https://codereview.chromium.org/573403002/diff/40001/components/metrics/metri...
File components/metrics/metrics_service_client.h (right):

https://codereview.chromium.org/573403002/diff/40001/components/metrics/metri...
components/metrics/metrics_service_client.h:36: virtual int32 GetProduct() = 0;
According to basictypes.h, we should use int32_t instead of the
Chromium-specific int32 typedef.

https://codereview.chromium.org/573403002/diff/40001/components/metrics/proto...
File components/metrics/proto/chrome_user_metrics_extension.proto (right):

https://codereview.chromium.org/573403002/diff/40001/components/metrics/proto...
components/metrics/proto/chrome_user_metrics_extension.proto:34: optional int32
product = 10 [default = 0];
Are you sure that we want a signed int rather than an unsigned int?  (I don't
remember how much of a difference it makes, but worth thinking about.)

[Alexei Svitkine (slow), 2014-09-26 16:45:38]

Patchset #2 (id:60001) has been deleted

[Alexei Svitkine (slow), 2014-09-26 16:51:10]

https://codereview.chromium.org/573403002/diff/40001/components/metrics/metri...
File components/metrics/metrics_log.cc (right):

https://codereview.chromium.org/573403002/diff/40001/components/metrics/metri...
components/metrics/metrics_log.cc:130:
prefs::kStabilityIncompleteSessionEndCount, 0);
On 2014/09/25 18:29:55, Ilya Sherman wrote:
> nit: Looks like this and several other lines below might not need to wrap
> anymore.  However, I'd slightly prefer that you remove the "metrics::"
namespace
> references in a separate CL, to keep this one less visually cluttered.

Done. I've kept the changes in this CL though, since they're trivial.

https://codereview.chromium.org/573403002/diff/40001/components/metrics/metri...
File components/metrics/metrics_service_client.h (right):

https://codereview.chromium.org/573403002/diff/40001/components/metrics/metri...
components/metrics/metrics_service_client.h:36: virtual int32 GetProduct() = 0;
On 2014/09/25 18:29:55, Ilya Sherman wrote:
> According to basictypes.h, we should use int32_t instead of the
> Chromium-specific int32 typedef.

Done.

https://codereview.chromium.org/573403002/diff/40001/components/metrics/proto...
File components/metrics/proto/chrome_user_metrics_extension.proto (right):

https://codereview.chromium.org/573403002/diff/40001/components/metrics/proto...
components/metrics/proto/chrome_user_metrics_extension.proto:34: optional int32
product = 10 [default = 0];
On 2014/09/25 18:29:55, Ilya Sherman wrote:
> Are you sure that we want a signed int rather than an unsigned int?  (I don't
> remember how much of a difference it makes, but worth thinking about.)

Enums can definitely be negative in protos and I'm pretty sure int32 is the
exact match for the underlying type for the enum, so I think int32 is the right
thing to use here (no need for signed/unsigned conversion when assigning enum
constants to the field).

[Ilya Sherman, 2014-09-26 17:24:42]

LGTM, thanks.

[Alexei Svitkine (slow), 2014-09-29 13:37:24]

The CQ bit was checked by asvitkine@chromium.org

[commit-bot: I haz the power, 2014-09-29 13:37:36]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/573403002/100001

[commit-bot: I haz the power, 2014-09-29 14:33:29]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2014-09-29 14:33:30]

Try jobs failed on following builders:
  android_clang_dbg_recipe on tryserver.chromium.linux
(http://build.chromium.org/p/tryserver.chromium.linux/builders/android_clang_d...)

[Alexei Svitkine (slow), 2014-09-29 20:35:57]

The CQ bit was checked by asvitkine@chromium.org

[commit-bot: I haz the power, 2014-09-29 20:37:14]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/573403002/100001

[commit-bot: I haz the power, 2014-09-29 20:57:41]

Committed patchset #3 (id:100001) as b794aa919a1da0af5baac40fb9d86e94b8bec00d

[commit-bot: I haz the power, 2014-09-29 20:58:32]

Patchset 3 (id:??) landed as
https://crrev.com/4c1d1ef9c811b58e83bff3e9cd8012d801bfa88b
Cr-Commit-Position: refs/heads/master@{#297254}