Extractlist Transform

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

{ "red":"100","white":"1300","blue":"315","purple":"55"}

The following step extracts the raw inventory contents of each color:

extractlist col: myWidgetInventory on:`{digit}+`

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

["100","1300","315","55"]

extractlist: col:column_ref on:string_literal_pattern delimiter:string_literal_pattern [quote:'quoted_string'] [as:'new_column_name']

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:

Identifies the pattern to match, which can be a string literal, Wrangle , or regular expression.

For the extractlist tranform, all instances that match this pattern in the source column are extracted into the array list in the new column. Each occurrence in the generated array corresponds to an individual instance in the source; the new column can contain duplicate values.

To create array elements based only on the delimiter parameter, set the following regular expression:

on:`/+/`

Usage Notes:

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 Wrangle.

In the following:

{ key1=value1,key2=value2 }

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

For this transform, this parameter defines the pattern that separates the values that you want to extract into the array.

Usage Notes:

extractlist col: MySourceValues on:`{alpha}+` delimiter:';' quote:'\"'

Output: Extracts from the MySourceValues column each instance of a string value that occurs before the delimiter. Values between double-quotes are considered string literals and are not processed according to the delimiters defined in the transform.

Can be used to specify a string as a single quoted object. This parameter value can be one or more characters.

Usage Notes:

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

Usage Notes:

In this example, you extract one or more values from a source column and assemble them in an Array column.

Suppose you need to extract the hashtags from customer tweets to another column. In such cases, you can use the {hashtag} Alteryx pattern to extract all hashtag values from a customer's tweets into a new column.

Source:

The following dataset contains customer tweets across different locations.

Transformation:

The following transformation extracts the hashtag messages from customer tweets.

customer_tweets

`{hashtag}`

Hashtag tweets

Then, the source column can be deleted.

Results:

Source:

In this example, a list of URLs identifies the items in the shopping carts of visitors to your web site. You want to extract the shopping cart information embedded in the query parameters of the URL.

Transformation:

The following transformation extracts the list of query values from the URL. Note that the equals sign is included in the matching pattern so that you don't accidentally pick up numeric values from the non-parameter part of the URL:

cartURL

`=[digit]+`

The two query parameter values have been extracted into an array of values, including the equals sign, which must be removed:

cartURL

'='

''

true

You can now unnest these values into separate columns:

extractlist_cartURL

'[0]','[1]'

After you rename the two columns to prodId and Qty, you can delete the column generated by the first transformation.

Results:

This example illustrates how to extract values from a column.

Source:

The following dataset contains counts of support emails processed by each member of the support team for individual customers over a six-month period. In this case, you are interested in the total number of emails processed for each customer.

Unfortunately, the data is ragged, as there are no entries for a support team member if he or she has not answered an email for a customer.

If the data is imported from a CSV file, you might need to make some simple Replace Text or Pattern transformations to clean up the data to look like the above example.

Transformation:

Use the following transformation to extract just the numeric values from the supportEmailCount array:

supportEmailCount

`{digit}+`

You should now have a column extractlist_supportEmailCount containing a ragged array. You can use the following transformations to convert this data to a comma-separated list of values:

extractlist_supportEmailCount

`[`

''

true

extractlist_supportEmailCount

`]`

''

true

extractlist_supportEmailCount

`"`

''

true

Convert the column to String data type.

You can now split out the column into separate columns containing individual values in the modified source. The limit parameter specifies the number of splits to create, resulting in 5 new columns, which is the maximum number of entries in the source arrays.

extractlist_supportEmailCount

On pattern

','

4

You might have to set the type for each generated column to Integer. If you try to use a New Formula transformation to calculate the sum of all of the generated columns, it only returns values for the first row because the missing rows are null values.

In the columns containing null values, select the missing value bar in the data histogram. Select the Replace suggestion card, and modify the transformation to write a 0 in place of the null value, as follows:

extractlist_supportEmailCount3

'0'

ismissing([extractlist_supportEmailCount3])

Repeat this step for any other column containing null values.

You can now use the following to sum the values in the generated columns:

Single row formula

(extractlist_supportEmailCount1 + extractlist_supportEmailCount2 + extractlist_supportEmailCount3 + extractlist_supportEmailCount4 + extractlist_supportEmailCount5)

Results:

After renaming the generated column to totalSupportEmails and dropping the columns used to create it, your dataset should look like the following: