Select
Select points from a stream satisfying particular properties.
The functionality provided by this module is similar to built-in reducers such as max(fieldA), min(fieldB); the difference is that reducers would place a single specified field into the resulting data point, while subgraphs in the select module would forward the entire data point with all its fields.
Since all select methods are subgraphs, they can be used only in stream context.
max - sub
... | select.max -field fieldname | ...
| Options | Description | Required? |
|---|---|---|
field |
the field name to check | Yes |
by |
list of fields to group by | No, default: [] |
Only the point with the largest value for the field name provided will be
forwarded.
min - sub
... | select.min -field fieldname | ...
| Options | Description | Required? |
|---|---|---|
field |
the field name to check | Yes |
by |
list of fields to group by | No, default: [] |
Only the point with the smallest value for the field name provided will be
forwarded.
percentile - sub
... | select.percentile -field fieldname -p percentile | ...
| Options | Description | Required? |
|---|---|---|
field |
the field name to check | Yes |
percentile |
the expected percentile value | Yes |
by |
list of fields to group by | No, default: [] |
Only the point with the expected percentile value for the field name
provided will be forwarded.
median - sub
... | select.median -field fieldname | ...
| Options | Description | Required? |
|---|---|---|
field |
the field name to check | Yes |
by |
list of fields to group by | No, default: [] |
Only the point with the exact median value for the field name
provided will be forwarded.
top - sub
... | select.top -n number -by field | ...
| Options | Description | Required? |
|---|---|---|
n |
how many of the points with the highest values of byto keep |
Yes |
by |
name of the field to sort by before figuring out the top nto keep |
Yes |
limit |
the limit value to pass to sort if you happen to exceed the default limit | No, default: see sort documentation |
Sort the inbound stream by the by field and return the n points that have
highest value of the by field.
Note Since top uses sort to sort by the by field we drop the time field
from the individual points.
bottom - sub
... | select.bottom -n number -by field | ...
| Options | Description | Required? |
|---|---|---|
n |
how many of the points with the lowest values of byto keep |
Yes |
by |
name of the field to sort by before figuring out the bottom nto keep |
Yes |
limit |
the limit value to pass to sort if you happen to exceed the default limit | No, default: see sort documentation |
Sort the inbound stream by the by field and return the n points that have
lowest value of the by field.
Note Since top uses sort to sort by the by field we drop the time field
from the individual points.