Calculate survivorship with last updated time


You can use the last updated time on incoming records when calculating survivorship. If there are two or more records from the same incoming source that match, then in the event of a disagreement over a field (because multiple source records might provide different values for a field), only the field value on the most recently updated record will survive.

Example

An incoming source file contains the following records:

vid__v first_name__v last_name__v last_updated middle_name__v suffix__v
929812881345544223 James Smith 2017-09-28T16:25:06Z Robert  
929812881345544223 Jim Smith 2017-09-28T16:35:06Z    
929812881345544223 James Smith 2017-09-28T16:10:06Z   Sr

Results

The records contain different values for fields. The field value with the last updated time survives.

Null values

Note that the treatment of null values is determined by the advanced property called "feed.retain.null": "<true/false>".

In this example, the value is false so Network did not consider the null values on the second record (most recently updated) for survivorship. If the value was true, the null values would be considered, so the resulting record would not have values for Middle Name and Suffix.

Enable this function

Add an advanced property to the source subscription to identify the column name that contains the last updated time.

To define which column in an incoming .csv file contains the last update time:

  1. In the Admin console, click System Interfaces > Source Subscriptions.
  2. Create a new subscription, or select an existing subscription, and click Advanced Mode.
  3. In the Edit Module Properties dialog, add the following property:

    "native.record.date.[entity type]": "[attribute name]"

    For example, in a typical source subscription the property would be: "native.record.date.HCP": "last_updated" where last_updated is the name of a column in the incoming .csv file.

Supported date and time formats

The date and time values in the source .csv file must be in one of the following formats:

  • yyyy-MM-dd - For example, 2017-02-27
  • yyyy-MM-ddTHH:mm:ssZ - For example, 2017-02-27T13:37:36Z