Support negative sample values in PersistentHistogramMap.

Support negative sample values in PersistentHistogramMap.

BUG=605229
Committed: https://crrev.com/82b10a43c1a329c1b198fdeddae101ba0dc029c5
Cr-Commit-Position: refs/heads/master@{#388758}

[bcwhite, 2016-04-20 21:49:50]

bcwhite@chromium.org changed reviewers:
+ asvitkine@chromium.org

[Alexei Svitkine (slow), 2016-04-20 21:51:18]

https://codereview.chromium.org/1909673002/diff/20001/base/metrics/persistent...
File base/metrics/persistent_sample_map.h (right):

https://codereview.chromium.org/1909673002/diff/20001/base/metrics/persistent...
base/metrics/persistent_sample_map.h:90: enum : HistogramBase::Sample {
kAllSamples = -10000 };
Can we avoid having this constant completely? For example, having a different
function to return all samples?

[Alexei Svitkine (slow), 2016-04-20 21:51:55]

https://codereview.chromium.org/1909673002/diff/20001/base/metrics/persistent...
File base/metrics/persistent_sample_map.h (right):

https://codereview.chromium.org/1909673002/diff/20001/base/metrics/persistent...
base/metrics/persistent_sample_map.h:90: enum : HistogramBase::Sample {
kAllSamples = -10000 };
On 2016/04/20 21:51:18, Alexei Svitkine wrote:
> Can we avoid having this constant completely? For example, having a different
> function to return all samples?

Or a separate boolean parameter?

[bcwhite, 2016-04-21 00:23:27]

Patchset #3 (id:40001) has been deleted

[bcwhite, 2016-04-21 00:23:35]

Patchset #3 (id:60001) has been deleted

[bcwhite, 2016-04-21 00:47:04]

Patchset #3 (id:80001) has been deleted

[bcwhite, 2016-04-21 10:33:15]

https://codereview.chromium.org/1909673002/diff/20001/base/metrics/persistent...
File base/metrics/persistent_sample_map.h (right):

https://codereview.chromium.org/1909673002/diff/20001/base/metrics/persistent...
base/metrics/persistent_sample_map.h:90: enum : HistogramBase::Sample {
kAllSamples = -10000 };
On 2016/04/20 21:51:55, Alexei Svitkine wrote:
> On 2016/04/20 21:51:18, Alexei Svitkine wrote:
> > Can we avoid having this constant completely? For example, having a
different
> > function to return all samples?
> 
> Or a separate boolean parameter?

Sure, though the two parameters would generally be mutually exclusive.  I
suppose I could pass a pointer to a Sample and allow "nullptr" to mean "all". 
Nah...  Boolean works better.

Done.

Two functions would be pretty much complete duplicates of each other so I don't
think it's a good idea.

[Alexei Svitkine (slow), 2016-04-21 13:34:29]

lgtm

[bcwhite, 2016-04-21 13:35:08]

The CQ bit was checked by bcwhite@chromium.org

[commit-bot: I haz the power, 2016-04-21 13:35:17]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1909673002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1909673002/100001

[commit-bot: I haz the power, 2016-04-21 13:39:19]

Committed patchset #3 (id:100001)

[commit-bot: I haz the power, 2016-04-22 19:34:11]

Description was changed from

==========
Support negative sample values in PersistentHistogramMap.

BUG=605229
==========

to

==========
Support negative sample values in PersistentHistogramMap.

BUG=605229

Committed: https://crrev.com/82b10a43c1a329c1b198fdeddae101ba0dc029c5
Cr-Commit-Position: refs/heads/master@{#388758}
==========

[commit-bot: I haz the power, 2016-04-22 19:34:12]

Patchset 3 (id:??) landed as
https://crrev.com/82b10a43c1a329c1b198fdeddae101ba0dc029c5
Cr-Commit-Position: refs/heads/master@{#388758}