Contents x
      CTOD
      • 2 Minutes to read
      • Print
      • Dark
        Light
      • PDF
      Contents

      CTOD

      • 2 Minutes to read
      • Print
      • Dark
        Light
      • PDF
      Article summary

      String: CTOD

      Attempts to create a new discrete field, usually by having applied a filter to a continuous string, integer or double.

      Purpose

      Use this function to:
      • Create a discrete field from a continuous field
      • Create a copy of a continuous field for use as a dimension in a profile
      • Create a copy of a field for use in a join
      • Turn a C (continuous) into a D (discrete)
      Processing Time
      CTOD can take a long time to run (i.e., upwards of 1 minute) if the following conditions are met:
      • Large number of records on the table (> 50 million)
      • Large number of discrete values (>300,000)
      CTOD will fail if the number of discrete values in field A exceeds the threshold which is 300,000 by default.     The threshold can be increased by running a BuildBakedField method in script editor and setting MAX_ENTRIES up to 2 million.   However, this can increase processing time significantly, so should only be done where necessary.  

      Return Value


      PropertyValue
      FieldTypeString  / Integer / Long(Integer) / Double 
      FieldSizeLow / Med / High 
      DataTypeDiscrete 
      DataSizeShort / Integer / LongInteger / Double
      Example Return Value: Return values are unchanged from Input value.   CTOD is not applicable for Dates or DateTimes.

      CTOD_Sample


      Parameters

      ParameterJSONDescription
      Table“targetTable”: “MyTableName”The target table on which the new field will be created
      Filter“dataset”: {DataSet_JSON}Optional.   If a filter is applied, records not in the filter recordset will be returned as null.
      Function"function":""String
      String Function“p1”: “CTOD(A)”CTOD(A)
      A“p2”:Required.   The discrete field to copy as a continuous field.
      Supports:  
      • DataType: Discrete
      • Field Type: Integer, Double, String
      MAX_ENTRIES
      		The threshold for crossing from discrete data type to continuous data type.

      JSON Sample

      {
  "method": "BuildBakedField",
  "targetTable": "BikeTripData",
  "overwrite": true,
  "name": "Journey_Random2",
  "function": "string",
  "p1": "CTOD(A)",
  "p2": "Journey_ID",
  "MAX_ENTRIES": 1000000,
  "project": "Model1"
}


      Usage Notes

      On very large tables, it may be necessary to use a filter in order to extract a discrete field from a continuous field.  In the example below, a random sampe of size 800,000 has been applied to the BikeTripData table in order to create the Journey_Random field. 

      CTOD_Example1


      Another approach is to use the BuildBakedField method and extend the MAX_ENTRIES threshold.   This approach should be used with care, as DataJet's threshold limits are designed to maximize performance across the application.  Having a large number of very high cardinality indexed fields (i.e., string fields above 200,000 unique values) may result in a perceptible (i.e., greater than 1 second) response time for operations such as populating the Context Panel grid.

      <<TODO:   Analyse field function >>

      CTOD_Example2

      See Also:   

      Example

      ExampleDetails
      DescriptionConvert a continuous string field to a discrete string
      Input
      • Table: = [BikeTripData]
      • Filter: = [No Filter]
      • Function: = String
      • String Function: = CTOD(A)
      • A:= [BikeTripData].[JourneyID]
      • Name: = Journey_Random2
      SampleCTOD_Example3




      Was this article helpful?
      What's Next