Available: coalesce() v1.159.0
The coalesce()
function is available from version
1.159.0.
Accepts a list of fields and returns the first value that is not null or
empty. For example, when accessing a value that may exist in any number of
potential fields, a case
statement could be used to select between the fields to identify the value
from the collection of fields. Using coalesce()
returns the first matching value across the selection of supplied fields.
If empty string values should be returned instead of being ignored, the
ignoreEmpty
parameter can be
set to false
to change the behavior.
Parameter | Type | Required | Default Value | Description |
---|---|---|---|---|
as | string | optional[a] | _coalesce | The field that contains the selected value, if any non-null value is found. |
expressions [b] | list of expressions | required | The list of expression candidates to select from; the first non-null result from left to right is used. | |
ignoreEmpty | boolean | optional[a] | true | If true , empty strings are treated as undefined values, that is, they are not selected. |
[a] Optional parameters use their default value unless explicitly set. [b] The parameter name |
Hide omitted argument names for this function
Omitted Argument NamesThe argument name for
expressions
can be omitted; the following forms of this function are equivalent:logscale Syntaxcoalesce("value")
and:
logscale Syntaxcoalesce(expressions="value")
These examples show basic structure only.
coalesce()
Examples
Click
next to an example below to get the full details.Find the First Values in a List of Fields
Find the first values of a list of fields to normalize data using the coalesce()
function
Query
coalesce([host, server, host[0].name, "example.com"])
Introduction
The coalesce()
function accepts a list of
fields and returns the first value that is not null or empty.
Using coalesce()
returns the first matching
value across the selection of supplied fields. In this example,
coalesce()
is used to normalize data from
different sources — the fields have the same meaning but
different names in the input data.
Example incoming data might look like this:
host='' |
server='crowdstrike.com' |
host[0].name='crowdstrike.com' |
machine='clienta' |
Step-by-Step
Starting with the source repository events.
- logscale
coalesce([host, server, host[0].name, "example.com"])
Finds the values of the first three fields host, server, host[0].name and the value of a string
"example.com"
and returns the results in a new field named __coalesce. Notice that the query uses a string literal as the last expression, which serves as a default value, because its value is not null. The first three expressions, on the other hand, are field names.In this example, the field names are simple and do not contain unsupported characters. If the field names contain unsupported characters, for example a space or an operator like ‘-', then the field cannot be quoted in Falcon LogScale, as it would be interpreted as string literals, in these situations, the
getField()
must be used together with thecoalesce()
function:coalesce([getField("host-name"), getField("server name"), "example.com"])
Event Result set.
Summary and Results
The query is used to normalize data from different sources by
finding the first value of a list of fields that are defined.
The coalesce()
function is useful if, for
example, you want to easily pick the first non-null value from
the list of prioritized fields and save it as a new field, or if
you want to be able to use default (string) value or an
expression instead of field name as an argument.
Sample output from the incoming example data:
_coalesce | host | server | host[0].name | machine |
---|---|---|---|---|
crowdstrike.com | <no value> | crowdstrike.com |