Page tree

 

Support | BlogContact Us | 844.332.2821

 

Contents:

This documentation applies to Trifacta Wrangler. Download this free product.
Registered users of this product or Trifacta Wrangler Enterprise should login to Product Docs through the application.

Contents:


Replaces values in the specified column or columns with the specified value, which can be a literal or an expression. Expressions can use conditional functions to filter the set of rows.

The set transform is used to replace entire cell values. For replacement of partial cell values using literals or patterns, use the replace transform. See Replace Transform.

Basic Usage

Literal example: 

set col: Country value: 'USA'

Output: Sets the values of all rows in the Country column to USA

Multi-column Literal example: 

set col: SSN,Phone value: '##REDACTED###'

Output: Sets the values of all rows in the SSN and Phone columns to ##REDACTED##

Expression example:

set col: isAmerica value: IF(Country == 'USA', true', 'false')

Output: If the value in the Country column is USA, then the value in isAmerica is set to true

Placeholder example:

You can substitute a placeholder value for the column name, which is useful if you are applying the same function across multiple columns. For example:

set col:score1,score2 value:IF ($col == 0, AVERAGE($col), $col)

Output: In the above transform, the values in score1 and score2 are set to the average of the column value when the value in the column is 0. Note that the computation of average is applied across all rows in the column, instead of just the filtered rows.

Window function example:

You can use window functions in your set transforms:

set col: avgSales value: ROLLINGAVERAGE(POS_Sales, 7, 0) group: saleDate order: saleDate

Output: Calculate the value in the column of avgSales to be the rolling average of the POS_Sales values for the preceding seven days, grouped and ordered by the saleDate column. For more information, see Window Functions.

Parameters

set col:col1,[col2] value:(expression) [group: group_col] 

ParameterRequired?Tranform BuilderData TypeDescription
col1YColumnstringColumn name
col2NColumnstringColumn name
valueYFormulastringExpression that generates the value to store in the column
groupNGroup bystringIf you are using aggregate or window functions, you can specify a group expression to identify the subset of records to apply the value expression.

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

col1, col2

Identifies the column and optional additional columns to which to apply the transform.

set col: MyCol value: 'myNewString'

Output: Sets value in  MyCol column to myNewString.

Usage Notes:

Required?Data Type
YesString (column name)

value

Identifies the expression that is applied by the transform. The value parameter can be one of the following types:

  • test predicates that evaluate to Boolean values (value: myAge == '30' yields a true or false value), or 
  • computational expressions ( value: abs(pow(myCol,3)) ).

The expected type of value expression is determined by the transform type. Each type of expression can contain combinations of the following:

  • literal values: value: 'Hello, world'
  • column references: value: amountOwed * 10
  • functions:  value: left(myString, 4)

  • combinations: value: abs(pow(myCol,3))

The types of any generated values are re-inferred by the platform. 

Usage Notes:

Required?Data Type
YesString (literal, column name, or expression)

group

Identifies the column by which the dataset is grouped for purposes of applying the transform.

If the value parameter contains aggregate or window functions, you can apply the group parameter to specify subsets of records across which the value computation is applied. 

You can specify one or more columns by which to group using comma-separated column references. 

Usage Notes:

Required?Data Type
NoString (column name)


Examples

Example - Clean up marketing contact data with replace, set, and extract

This example illustrates the different uses of the following transforms to replace or extract cell data:

  • set - defines the values to use in a predefined column. See Set Transform.

    Tip: Use the derive transform to generate a new column containing a defined set of values. See Derive Transform.

  • replace - replaces a string literal or pattern appearing in the values of a column with a specific string. See Replace Transform.
  • extract - extracts a pattern-based value from a column and stores it in a new column. See Extract Transform.

Source:

The following dataset contains contact information that has been gathered by your marketing platform from actions taken by visitors on your website. You must clean up this data and prepare it for use in an analytics platform.

LeadIdLastNameFirstNameTitlePhoneRequest
LE160301001JonesCharlesChief Technical Officer415-555-1212reg
LE160301002LyonsEdward 415-012-3456download whitepaper
LE160301003MartinMaryCEO510-555-5555delete account
LE160301004SmithTaliaEngineer510-123-4567free trial

 

Transform:

Title column: For example, you first notice that some data is missing. Your analytics platform recognizes the string value, "#MISSING#" as an indicator of a missing value. So, you click the missing values bar in the Title column. Then, you select the Replace suggestion card. Note that the default replacement is a null value, so you click Modify and update it:

set col: Title value: IF(ISMISSING([Title]),'#MISSING#',Title)

Request column: In the Request column, you notice that the reg entry should be cleaned up. Add the following transform, which replaces that value:

replace col:Request with:'Registration' on:`{start}reg{end}`

The above transform uses a Trifacta® pattern as the expression of the on: parameter. This expression indicates to match from the start of the cell value, the string literal reg, and then the end of the cell value, which matches on complete cell values of reg only.

This transform works great on the sample, but what happens if the value is Reg with a capital R? That value might not be replaced. To improve the transform, you can modify the transform with the following Trifacta pattern in the on parameter, which captures differences in capitalization:

replace col:Request with:'Registration' on:`{start}{[R|r]}eg{end}`

Add the above transform to your recipe. Then, it occurs to you that all of the values in the Request column should be capitalized in title or proper case:

set col:Request value:PROPER(Request)

Now, all values are capitalized as titles.

Phone column: You might have noticed some issues with the values in the Phone column. In the United States, the prefix 555 is only used for gathering information; these are invalid phone numbers.

In the data grid, you select the first instance of 555 in the column. However, it selects all instances of that pattern, including ones that you don't want to modify. In this case, continue your selection by selecting the similar instance of 555 in the other row. In the suggestion cards, you click the Replace transform.

Notice, however, that the default Replace transform has also highlighted the second 555 pattern in one instance, which could be a problem in other phone numbers not displayed in the sample. You must modify the selection pattern for this transform. In the on: parameter below, the Trifacta pattern has been modified to match only the instances of 555 that appear in the second segment in the phone number format:

replace col: Phone on: `{start}%{3}-555-%*{end}` with: '#INVALID#' global: true

Note the wildcard construct has been added (%*). While it might be possible to add a pattern that matches on the last four characters exactly (%{4}), that matching pattern would not capture the possibility of a phone number having an extension at the end of it. The above expression does.

NOTE: The above transform creates values that are mismatched with the Phone Number data type. In this example, however, these mismatches are understood to be for the benefit of the system consuming your Trifacta output.

LeadId column: You might have noticed that the lead identifier column (LeadId) contains some embedded information: a date value and an identifier for the instance within the day. The following steps can be used to break out this information. The first one creates a separate working column with this information, which allows us to preserve the original, unmodified column:

derive value:LeadId as:'LeadIdworking'

You can now work off of this column to create your new ones. First, you can use the following replace transform to remove the leading two characters, which are not required for the new columns:

replace col:LeadIdworking with:'' on:'LE'

Notice that the date information is now neatly contained in the first characters of the working column. Use the following to extract these values to a new column:

extract col: LeadIdworking on: `{start}%{6}`

The new LeadIdworking2 column now contains only the date information. Cleaning up this column requires reformatting the data, retyping it as a Datetime type, and then applying the dateformat function to format it to your satisfaction. These steps are left as a separate exercise.

For now, let's just rename the column:

rename col:LeadIdworking1 to:'LeadIdDate'

In the first working column, you can now remove the date information using the following:

replace col: LeadIdworking on: `{start}%{6}` with: ''

You can rename this column to indicate it is a daily identifier:

rename col:LeadIdworking to:'LeadIdDaily'

 

Results:

LeadIdLeadIdDailyLeadIdDateLastNameFirstNameTitlePhoneRequest
LE160301001001160301JonesCharlesChief Technical Officer#INVALID#Registration
LE160301002002160301LyonsEdward#MISSING#415-012-3456Download Whitepaper
LE160301003003160301MartinMaryCEO#INVALID#Delete Account
LE160301004004160301SmithTaliaEngineer510-123-4567Free Trial

 

 

Example - Using $col placeholder

This example illustrates how you can use the following conditional calculation functions to analyze weather data:

  • AVERAGEIF - Average of a set of values by group that meet a specified condition. See AVERAGEIF Function.
  • MINIF - Minimum of a set of values by group that meet a specified condition. See MINIF Function.
  • MAXIF - Maximum of a set of values by group that meet a specified condition. See MAXIF Function.
  • VARIF - Variance of a set of values by group that meet a specified condition. See VARIF Function.
  • STDEVIF - Standard deviation of a set of values by group that meet a specified condition. See STDEVIF Function.

Source:

Here is some example weather data:

datecityrain_cmtemp_Cwind_mph
1/23/17Valleyville0.0012.86.7
1/23/17Center Town0.319.45.3
1/23/17Magic Mountain0.000.07.3
1/24/17Valleyville0.2517.23.3
1/24/17Center Town0.541.17.6
1/24/17Magic Mountain0.325.08.8
1/25/17Valleyville0.023.36.8
1/25/17Center Town0.833.35.1
1/25/17Magic Mountain0.59-1.76.4
1/26/17Valleyville1.0815.04.2
1/26/17Center Town0.966.17.6
1/26/17Magic Mountain0.77-3.93.0
1/27/17Valleyville1.007.22.8
1/27/17Center Town1.3220.00.2
1/27/17Magic Mountain0.775.65.2
1/28/17Valleyville0.12-6.15.1
1/28/17Center Town0.145.04.9
1/28/17Magic Mountain1.501.10.4
1/29/17Valleyville0.3613.37.3
1/29/17Center Town0.756.19.0
1/29/17Magic Mountain0.603.36.0


Transform:

The following computes average temperature for rainy days by city:

derive value:AVERAGEIF(temp_C, rain_cm > 0) group:city as:'avgTempWRain'

The following computes maximum wind for sub-zero days by city:

derive value:MAXIF(wind_mph,temp_C < 0) group:city as:'maxWindSubZero'

This step calculates the minimum temp when the wind is less than 5 mph by city:

derive value:MINIF(temp_C,wind_mph<5) group:city as:'minTempWind5'

This step computes the variance in temperature for rainy days by city:

derive value:VARIF(temp_C,rain_cm >0) group:city as:'varTempWRain'

The following computes the standard deviation in rainfall for Center Town:

derive value:STDEVIF(rain_cm,city=='Center Town') as:'stDevRainCenterTown'

You can use the following transforms to format the generated output. Note the $col placeholder value for the multi-column transforms:

set col:stDevRainCenterTown,maxWindSubZero value:numformat($col,'##.##')

Since the following rely on data that has only one significant digit, you should format them differently:

set col:varTempWRain,avgTempWRain,minTempWind5 value:numformat($col,'##.#')

 


Results:

Here is some example weather data:

datecityrain_cmtemp_Cwind_mphavgTempWRainmaxWindSubZerominTempWind5varTempWRainstDevRainCenterTown
1/23/17Valleyville0.0012.86.78.35.17.263.80.37
1/23/17Center Town0.319.45.37.3 532.60.37
1/23/17Magic Mountain0.000.07.31.66.43-3.9120.37
1/24/17Valleyville0.2517.23.38.35.17.263.80.37
1/24/17Center Town0.541.17.67.3 532.60.37
1/24/17Magic Mountain0.325.08.81.66.43-3.9120.37
1/25/17Valleyville0.023.36.88.35.17.263.80.37
1/25/17Center Town0.833.35.17.3 532.60.37
1/25/17Magic Mountain0.59-1.76.41.66.43-3.9120.37
1/26/17Valleyville1.0815.04.28.35.17.263.80.37
1/26/17Center Town0.966.17.67.3 532.60.37
1/26/17Magic Mountain0.77-3.93.01.66.43-3.9120.37
1/27/17Valleyville1.007.22.88.35.17.263.80.37
1/27/17Center Town1.3220.00.27.3 532.60.37
1/27/17Magic Mountain0.775.65.21.66.43-3.9120.37
1/28/17Valleyville0.12-6.15.18.35.17.263.80.37
1/28/17Center Town0.145.04.97.3 532.60.37
1/28/17Magic Mountain1.501.10.41.66.43-3.9120.37
1/29/17Valleyville0.3613.37.38.35.17.263.80.37
1/29/17Center Town0.756.19.07.3 532.60.37
1/29/17Magic Mountain0.603.36.01.66.43-3.9120.37



 

Your Rating: Results: PatheticBadOKGoodOutstanding! 3 rates

This page has no comments.