int() argument must be a string, a bytes-like object or a real number, not 'NoneType'

[M39913]

I tried a multiple things, see below. But I keep getting this error:

int() argument must be a string, a bytes-like object or a real number, not 'NoneType'

in macro default__get_intervals_between (macros_utils\date_spine.sql)
called by macro get_intervals_between (macros_utils\date_spine.sql)
called by macro default__date_spine (macros_utils\date_spine.sql)
called by macro date_spine (macros_utils\date_spine.sql)
called by macro default__get_base_dates (macros\get_base_dates.sql)
called by macro get_base_dates (macros\get_base_dates.sql)
called by model Datum_tabel (models\scd0\Datum_tabel.sql)

packages:

• package: dbt-labs/dbt_utils
  version: 1.1.1
• package: calogica/dbt_date
  version: 0.10.1

I also tried previous versions.

The model I runned:
{{ dbt_date.get_base_dates(start_date="1960-01-01", end_date="2049-12-31")}}

I also tried changing the model to this:
{{ dbt_date.get_base_dates(n_dateparts=7) }}

And this:
{{ dbt_date.get_date_dimension("1990-01-01", "2050-12-31") }}

I use databricks version 1.7.0 and dbt version 1.7.7
Please let me know your thoughts and any possible solutions :)

[clausherther]

I was able to run both

{{ dbt_date.get_date_dimension("1990-01-01", "2050-12-31") }}

and

{{ dbt_date.get_base_dates(n_dateparts=7) }}

against BigQuery without issues. We don't really support databricks, so I can't tell you if it has to do with that adapter.
We do run integration tests against dbt-spark and run

dbt-date/integration_tests/models/dates.sql

Line 6 in c596b56

{{ dbt_date.get_date_dimension('2015-01-01', '2022-12-31') }}

and various flavors of

dbt-date/integration_tests/models/dim_hour.sql

Line 7 in c596b56

{{ dbt_date.get_base_dates(n_dateparts=24*28, datepart="hour") }}

in our test date models.

[clausherther]

Best thing would probably be to get the exact compiled run-time SQL line that's failing.

[M39913]

Thanks for your response! I tried the tests, both of them didn't work, I got the same error.

This is the complete error by the way:
Runtime Error
Compilation Error in model Datum_tabel (models\scd0\Datum_tabel.sql)
int() argument must be a string, a bytes-like object or a real number, not 'NoneType'

> in macro default__get_intervals_between (macros\_utils\date_spine.sql)
> called by macro get_intervals_between (macros\_utils\date_spine.sql)
> called by macro default__date_spine (macros\_utils\date_spine.sql)
> called by macro date_spine (macros\_utils\date_spine.sql)
> called by macro default__get_base_dates (macros\get_base_dates.sql)
> called by macro get_base_dates (macros\get_base_dates.sql)
> called by model Datum_tabel (models\scd0\Datum_tabel.sql)

So I get the error before he compiles.
I think it has to do something with macro default__get_intervals_between

[clausherther]

You'll need to find the actual SQL statement that fails, not the dbt macros.

get_intervals_between just calls dbt.datediff(start_date, end_date, datepart) but I don't know what dbt translates that to on databricks. You'll want to find what SQL this actually runs on your platform to see why it fails.

{%- call statement('get_intervals_between', fetch_result=True) %}

        select {{ dbt.datediff(start_date, end_date, datepart) }}

    {%- endcall -%}