Skip to main content

Flatten Transform

注意

Transforms are a part of the underlying language, which is not directly accessible to users. This content is maintained for reference purposes only. For more information on the user-accessible equivalent to transforms, see Transformation Reference.

Unpacks array data into separate rows for each value. This transform operates on a single column.

This transform does not reference keys in the array. If your array data contains keys, use the unnest transform. See Unnest Transform.

Basic Usage

flatten col: myArray

Output: Generates a separate row for each value in the array. Values of other columns in generated rows are copied from the source.

Syntax and Parameters

flatten: col: column_ref

Token

Required?

Data Type

Description

flatten

Y

transform

Name of the transform

col

Y

string

Source column name

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

col

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

Usage Notes:

Required?

Data Type

Yes

String (column name)

Examples

提示

For additional examples, see Common Tasks.

Example - Flatten an array

This section describes how to flatten the values in an Array into separate rows in your dataset.

Source:

In the following example dataset, students took the same test three times, and their scores were stored in any array in the Scores column.

LastName

FirstName

Scores

Adams

Allen

[81,87,83,79]

Burns

Bonnie

[98,94,92,85]

Cannon

Chris

[88,81,85,78]

Transformation:

When the data is imported, you might have to re-type the Scores column as an array:

Transformation Name

Change column data type

Parameter: Columns

Scores

Parameter: New type

Array

You can now flatten the Scores column data into separate rows:

Transformation Name

Expand Array into rows

Parameter: Column

Scores

Results:

LastName

FirstName

Scores

Adams

Allen

81

Adams

Allen

87

Adams

Allen

83

Adams

Allen

79

Burns

Bonnie

98

Burns

Bonnie

94

Burns

Bonnie

92

Burns

Bonnie

85

Cannon

Chris

88

Cannon

Chris

81

Cannon

Chris

85

Cannon

Chris

78

提示

You can use aggregation functions on the above data to complete values like average, minimum, and maximum scores. When these aggregation calculations are grouped by student, you can perform the calculations for each student.

Example - Flatten and unnest together

While the above example nicely flattens out your data, there are two potential problems with the results:

  • There is no identifier for each test. For example, Allen Adams' score of 87 cannot be associated with the specific test on which he recorded the score.

  • There is no unique identifier for each row.

The following example addresses both of these issues. It also demonstrates differences between the unnest and the flatten transform, including how you use unnest to flatten array data based on specified keys.

This example illustrates you to use the flatten and unnest transforms.

Source:

You have the following data on student test scores. Scores on individual scores are stored in the Scores array, and you need to be able to track each test on a uniquely identifiable row. This example has two goals:

  1. One row for each student test

  2. Unique identifier for each student-score combination

LastName

FirstName

Scores

Adams

Allen

[81,87,83,79]

Burns

Bonnie

[98,94,92,85]

Cannon

Charles

[88,81,85,78]

Transformation:

When the data is imported from CSV format, you must add a header transform and remove the quotes from the Scores column:

Transformation Name

Rename column with row(s)

Parameter: Option

Use row(s) as column names

Parameter: Type

Use a single row to name columns

Parameter: Row number

1

Transformation Name

Replace text or pattern

Parameter: Column

colScores

Parameter: Find

'\"'

Parameter: Replace with

''

Parameter: Match all occurrences

true

Validate test date: To begin, you might want to check to see if you have the proper number of test scores for each student. You can use the following transform to calculate the difference between the expected number of elements in the Scores array (4) and the actual number:

Transformation Name

New formula

Parameter: Formula type

Single row formula

Parameter: Formula

(4 - arraylen(Scores))

Parameter: New column name

'numMissingTests'

When the transform is previewed, you can see in the sample dataset that all tests are included. You might or might not want to include this column in the final dataset, as you might identify missing tests when the recipe is run at scale.

Unique row identifier: The Scores array must be broken out into individual rows for each test. However, there is no unique identifier for the row to track individual tests. In theory, you could use the combination of LastName-FirstName-Scores values to do so, but if a student recorded the same score twice, your dataset has duplicate rows. In the following transform, you create a parallel array called Tests, which contains an index array for the number of values in the Scores column. Index values start at 0:

Transformation Name

New formula

Parameter: Formula type

Single row formula

Parameter: Formula

range(0,arraylen(Scores))

Parameter: New column name

'Tests'

Also, we will want to create an identifier for the source row using the sourcerownumber function:

Transformation Name

New formula

Parameter: Formula type

Single row formula

Parameter: Formula

sourcerownumber()

Parameter: New column name

'orderIndex'

One row for each student test: Your data should look like the following:

LastName

FirstName

Scores

Tests

orderIndex

Adams

Allen

[81,87,83,79]

[0,1,2,3]

2

Burns

Bonnie

[98,94,92,85]

[0,1,2,3]

3

Cannon

Charles

[88,81,85,78]

[0,1,2,3]

4

Now, you want to bring together the Tests and Scores arrays into a single nested array using the arrayzip function:

Transformation Name

New formula

Parameter: Formula type

Single row formula

Parameter: Formula

arrayzip([Tests,Scores])

Your dataset has been changed:

LastName

FirstName

Scores

Tests

orderIndex

column1

Adams

Allen

[81,87,83,79]

[0,1,2,3]

2

[[0,81],[1,87],[2,83],[3,79]]

Adams

Bonnie

[98,94,92,85]

[0,1,2,3]

3

[[0,98],[1,94],[2,92],[3,85]]

Cannon

Charles

[88,81,85,78]

[0,1,2,3]

4

[[0,88],[1,81],[2,85],[3,78]]

Use the following to unpack the nested array:

Transformation Name

Expand arrays to rows

Parameter: Column

column1

Each test-score combination is now broken out into a separate row. The nested Test-Score combinations must be broken out into separate columns using the following:

Transformation Name

Unnest Objects into columns

Parameter: Column

column1

Parameter: Paths to elements

'[0]','[1]'

After you delete column1, which is no longer needed you should rename the two generated columns:

Transformation Name

Rename columns

Parameter: Option

Manual rename

Parameter: Column

column_0

Parameter: New column name

'TestNum'

Transformation Name

Rename columns

Parameter: Option

Manual rename

Parameter: Column

column_1

Parameter: New column name

'TestScore'

Unique row identifier: You can do one more step to create unique test identifiers, which identify the specific test for each student. The following uses the original row identifier OrderIndex as an identifier for the student and the TestNumber value to create the TestId column value:

Transformation Name

New formula

Parameter: Formula type

Single row formula

Parameter: Formula

(orderIndex * 10) + TestNum

Parameter: New column name

'TestId'

The above are integer values. To make your identifiers look prettier, you might add the following:

Transformation Name

Merge columns

Parameter: Columns

'TestId00','TestId'

Extending: You might want to generate some summary statistical information on this dataset. For example, you might be interested in calculating each student's average test score. This step requires figuring out how to properly group the test values. In this case, you cannot group by the LastName value, and when executed at scale, there might be collisions between first names when this recipe is run at scale. So, you might need to create a kind of primary key using the following:

Transformation Name

Merge columns

Parameter: Columns

'LastName','FirstName'

Parameter: Separator

'-'

Parameter: New column name

'studentId'

You can now use this as a grouping parameter for your calculation:

Transformation Name

New formula

Parameter: Formula type

Single row formula

Parameter: Formula

average(TestScore)

Parameter: Group rows by

studentId

Parameter: New column name

'avg_TestScore'

Results:

After you delete unnecessary columns and move your columns around, the dataset should look like the following:

TestId

LastName

FirstName

TestNum

TestScore

studentId

avg_TestScore

TestId0021

Adams

Allen

0

81

Adams-Allen

82.5

TestId0022

Adams

Allen

1

87

Adams-Allen

82.5

TestId0023

Adams

Allen

2

83

Adams-Allen

82.5

TestId0024

Adams

Allen

3

79

Adams-Allen

82.5

TestId0031

Adams

Bonnie

0

98

Adams-Bonnie

92.25

TestId0032

Adams

Bonnie

1

94

Adams-Bonnie

92.25

TestId0033

Adams

Bonnie

2

92

Adams-Bonnie

92.25

TestId0034

Adams

Bonnie

3

85

Adams-Bonnie

92.25

TestId0041

Cannon

Chris

0

88

Cannon-Chris

83

TestId0042

Cannon

Chris

1

81

Cannon-Chris

83

TestId0043

Cannon

Chris

2

85

Cannon-Chris

83

TestId0044

Cannon

Chris

3

78

Cannon-Chris

83