Skip to main content
Version: 2.3.0

Split

Split transform plugin

Description​

A string cutting function is defined, which is used to split the specified field in the Sql plugin.

tip

This transform both supported by engine Spark and Flink.

Options​

nametyperequireddefault value
separatorstringno" "
fieldsarrayyes-
source_fieldstringnoraw_message
target_fieldstringnoroot
common-optionsstringno-

separator [string]​

Separator, the input string is separated according to the separator. The default separator is a space (" ") . Note: If you use some special characters in the separator, you need to escape it. e.g. "\|"

source_field [string]​

The source field of the string before being split, if not configured, the default is raw_message

target_field [string]​

target_field can specify the location where multiple split fields are added to the Event. If it is not configured, the default is _root_ , that is, all split fields will be added to the top level of the Event. If a specific field is specified, the divided field will be added to the next level of this field.

fields [list]​

In the split field name list, specify the field names of each character string after splitting in order. If the length of the fields is greater than the length of the separation result, the extra fields are assigned null characters.

common options [string]​

Transform plugin common parameters, please refer to Transform Plugin for details

Examples​

Split the message field in the source data according to &, you can use field1 or field2 as the key to get the corresponding value

split {
source_field = "message"
separator = "&"
fields = ["field1", "field2"]
}

Split the message field in the source data according to , , the split field is info , you can use info.field1 or info.field2 as the key to get the corresponding value

split {
source_field = "message"
target_field = "info"
separator = ","
fields = ["field1", "field2"]
}

Use Split as udf in sql.

  # This just created a udf called split
Split{
separator = "#"
fields = ["name","age"]
}
# Use the split function (confirm that the fake table exists)
sql {
sql = "select * from (select raw_message,split(raw_message) as info_row from fake) t1"
}