0

Are there any specific rules for column names in great expectations? In particular, if you have a column like a.age ? would it have to be renamed to a_age in order to run an expectation on it?

stackguy1723
  • 165
  • 1
  • 2
  • 12

1 Answers1

0

The application of an expectation expects that it uses the name that the column assumes in the starting dataset. However you can put in the comments section a reference to what you mean. For example, in the case you mentioned:

{
      "expectation_type": "expect_column_values_to_not_be_null",
      "kwargs": {
        "column": "a.age"
      },
      "meta": {
        "notes": {
          "content": "a_age",
          "format": "markdown"
        }
      }
    }

and this you will find in the json file of the results. To answer your question: changing the name of the column by altering the dataset, with great_expectations, is not possible because one of the fundamental objectives is to apply expectations by making as few alterations as possible on the data.

Veronica
  • 1