Decrease minimal granularity about druidplugin

[iamsurya-dev]

Will take a look at this.

[pkarmaka]

We can change the custom granularity to include option for second, but lets keep the default as minute.

[RestfulBlue]

i think just add 2 missing druid granularities ( none and second ) will be good enough =)

[knoguchi]

none is rarely useful but if we add it why not other supported granularities as well? Per Druid documentation, the supported granularity strings are: all, none, second, minute, fifteen_minute, thirty_minute, hour, day, week, month, quarter and year.

[iamsurya-dev]

Agree that none is rarely useful. It was just something that I added to address this issue. We could add all other granularities as well. Let me know if you think I should add it. I don't see a use of quarter and year granularity from a grafana usecase though.

[RestfulBlue]

i think main problem using none granularity with grafana is empty buckets.
according to docs :
http://druid.io/docs/latest/querying/timeseriesquery.html
druid support possibility to skip empty buckets adding flag skipEmptyBuckets to context

{
"queryType": "timeseries",
"dataSource": "sample_datasource",
"granularity": "day",
"aggregations": [
{ "type": "longSum", "name": "sample_name1", "fieldName": "sample_fieldName1" }
],
"intervals": [ "2012-01-01T00:00:00.000/2012-01-04T00:00:00.000" ],
"context" : {
"skipEmptyBuckets": "true"
}
}
i think adding checkbox to add this flag will make it usable in many place. In some place we collect data with 1s interval. But due to some different lags, sometime 2 metrics can be writed with same second. To avoid visualization problem with second granularity we should add post aggregation. Also with that situation come empty bucket, as 2 metrics have some time, so chart look strange in some place. Using none granularity and skipEmptyBuckets will make chart look as it is, without empty buckets and additional post aggregation.