No return_expectations error for study with default_expectations

[inglesp]

Given the following study definition:

study = StudyDefinition(
    default_expectations={
        "date": {"earliest": "1900-01-01", "latest": "today"},
        "rate": "exponential_increase",
    },

    population=patients.registered_with_one_practice_between(
        "2019-02-01", "2020-02-01"
    ),

    asthma_dx_ever = patients.with_these_clinical_events(
        asthma_dx_codes
    ),
)

when I try to generate dummy data, I get this error:

     2021-02-23 15:24:19 [info     ] Generating cohort for study_definition in output [cohortextractor.cohortextractor] 
     Exception at 2021-02-23 15:24:19 UTC
     Traceback (most recent call last):
       File "/usr/local/bin/cohortextractor", line 33, in <module>
         sys.exit(load_entry_point('opensafely-cohort-extractor', 'console_scripts', 'cohortextractor')())
       File "/app/cohortextractor/cohortextractor.py", line 695, in main
         generate_cohort(
       File "/app/cohortextractor/cohortextractor.py", line 145, in generate_cohort
         _generate_cohort(
       File "/app/cohortextractor/cohortextractor.py", line 190, in _generate_cohort
         study.to_csv(
       File "/app/cohortextractor/study_definition.py", line 72, in to_csv
         df = self.make_df_from_expectations(expectations_population)
       File "/app/cohortextractor/study_definition.py", line 257, in make_df_from_expectations
         raise ValueError(f"No `return_expectations` defined for {colname}")
     ValueError: No `return_expectations` defined for asthma_dx_ever

I would expect the default_expectations to be used so that return_expectations is unnecessary. Am I doing something wrong?

I'm running opensafely v1.6.2.

[alexwalkerepi]

I think as asthma_dx_ever is a binary flag, it needs an expectation to tell it how frequently it should occur, e.g. "incidence": 0.5. The "rate" expectation is a bit misleading as it really relates just to the distribution of dates.

[inglesp]

Makes sense, thank you!