dbeatty10
commented
2 years ago Option X
Calling it for a known, explicit type would look like this:
{{ cast(expression, type_integer()) }}Calling it for a variable data type would look like this:
{{ cast(expression, api.Column.translate_type(data_type)) }}The default implementation might look like this:
-- core/dbt/include/global_project/macros/utils/cast.sql
{% macro cast(expression, data_type) %}
{{ adapter.dispatch('cast', 'dbt') (expression, data_type) }}
{% endmacro %}
{% macro default__cast(expression, data_type) -%}
cast({{ expression }} as {{ data_type }})
{%- endmacro %}Option Y
Calling it for an explicit type would look like this:
{{ cast(expression, "integer") }}Calling it for a variable data type would look like this:
{{ cast(expression, data_type) }}The default implementation would look like:
-- core/dbt/include/global_project/macros/utils/cast.sql
{% macro cast(expression, data_type) %}
{{ adapter.dispatch('cast', 'dbt') (expression, data_type) }}
{% endmacro %}
{% macro default__cast(expression, data_type) -%}
cast({{ expression }} as {{ api.Column.translate_type(data_type) }})
{%- endmacro %}
jtcohen6
alison985
Is this your first time submitting a feature request?
Describe the feature
Per conversation with @jtcohen6 we are interested in creating a
cast()macro meant to behave like theCASToperator described in the SQL standard.Usage likely to be something like:
Which some databases might render as:
Describe alternatives you've considered
Who will this benefit?
Like all cross-database macros, we anticipate this being most relevant to dbt package maintainers.
It would also be useful in implementing this:
Are you interested in contributing this feature?
Yep!
Anything else?
N/A