Extractkv Transform

Extracts key-value pairs from a source column and writes them to a new column.Source column must be of String type, although the data can be formatted as other data types. The generated column is of Object type.

Your source column (MyKeyValues) is formatted in the following manner:


Basic Usage

The following transform extracts the key-value pairs. The key parameter contains a single pattern that matches all keys that you want to extract:

extractkv col: MyKeyValues key:`{alpha}+{digit}` valueafter: '=' delimiter: ','

Output: The generated column contains data that looks like the following:


If the source data contained additional keys which were not specified in the transform, those key-value pairs would not appear in the generated column.


extractkv col:column_ref delimiter:string_literal_pattern key:string_literal_pattern valueafter:string_literal_pattern [as:'new_column_name']

ParameterRequired?Transform BuilderData TypeDescription
extractkvYConvert Key/Values into ObjectstransformName of the transform
colYColumnstringSource column name
delimiterYKey/Value pair delimiterstringString literal or pattern that identifies the separator between key-value pairs
keyYKey patternstringPattern that identifies the key to match
valueafterYKey Value separatorstringString literal or pattern after which is located a key's value
asNNew column namestringName of the newly generated column

For more information on syntax standards, see Language Documentation Syntax Notes.


Identifies the column to which to apply the transform. You can specify only one column.

Usage Notes:

Required?Data Type
YesString (column name)


Specifies the character or pattern that defines the end of a key-value pair. This value can be specified as a String literal, regular expression, or Cloud Dataprep pattern.

In the following:

{ key1=value1,key2=value2 }

The delimiter is the comma ( ','). The final key-value pair does not need a delimiter.

Tip: You can insert the Unicode equivalent character for this parameter value using a regular expression of the form /\uHHHH/. For example, /\u0013/ represents Unicode character 0013 (carriage return). For more information, see Supported Special Regular Expression Characters.

Usage Notes:

Required?Data Type

String (literal, regular expression, or Cloud Dataprep pattern)


Specifies the pattern used to extract the keys from a source column by the extractkv transform. For the following data:

{ key1=value1,key2=value2 }

The keys are represented in the transform by the following parameter and value:


This pattern matches all keys that begin with a letter and end with a digit. If the source data contains other keys, they do not appear in the extracted data.

Usage Notes:

Required?Data Type
YesSingle pattern representing the individual keys to extract.


Specifies the character or pattern after which the value is specified in a key-value pair. This value can be specified as a String literal, regular expression, or Cloud Dataprep pattern.

For the following:

{ key1=value1,key2=value2 }

The valueafter string is the equals sign ( '=').

Usage Notes:

Required?Data Type

String (literal, regular expression, or Cloud Dataprep pattern)


Name of the new column that is being generated. If the as parameter is not specified, a default name is used.

Usage Notes:

Required?Data Type
NoString (column name)


Example - extracting key values from car data and the unnesting into separate columns

This example shows how you can unpack data nested in an Object into separate columns using the following transforms:


You have the following information on used cars. The VIN column contains vehicle identifiers, and the Properties column contains key-value pairs describing characteristics of each vehicle. You want to unpack this data into separate columns.

XX3 JT4522year=2004,make=Subaru,model=Impreza,color=green,mileage=125422,cost=3199
HT4 UJ9122year=2006,make=VW,model=Passat,color=silver,mileage=102941,cost=4599
KC2 WZ9231year=2009,make=GMC,model=Yukon,color=black,mileage=68213,cost=12899
LL8 UH4921year=2011,make=BMW,model=328i,color=brown,mileage=57212,cost=16999


Add the following transform, which identifies all of the key values in the column as beginning with alphabetical characters.

  • The valueafter string identifies where the corresponding value begins after the key.
  • The delimiter string indicates the end of each key-value pair.

extractkv col:Properties key:`{alpha}+` valueafter:`=` delimiter:`,`

Now that the Object of values has been created, you can use the unnest transform to unpack this mapped data. In the following, each key is specified, which results in separate columns headed by the named key:

unnest col:extractkv_Properties keys:'year','make','model','color','mileage','cost'


When you drop the unnecessary Properties columns, the dataset now looks like the following:

XX3 JT45222004SubaruImprezagreen1254223199
HT4 UJ91222006VWPassatsilver1029414599
KC2 WZ92312009GMCYukonblack6821312899
LL8 UH49212011BMW328ibrown5721216999

Send feedback about...

Google Cloud Dataprep Documentation