Snowflake SQL compilation error with quoted column names

[yannp]

Hi,
I encounter an issue I don't know how to handle. I get a SQL compilation error when running a test on my model when this model has quoted column names (in this example using "first name" and "last name" columns defined in a dummy jaffleshop model). A simple unit test looks like this:

{% call dbt_unit_testing.test('jaffle', 'name_is_correct') %}
  {% call dbt_unit_testing.mock_ref('jaffleshop', {"input_format": "sql"}) %}
    select 'bob' as "first name", 'bobson' as "last name"
  {% endcall %}

  {% call dbt_unit_testing.expect({"input_format": "sql"}) %}
    select 'bob' as "first name"
  {% endcall %}
{% endcall %}

Output gives invalid identifier '"FIRST NAME"'.
Thanks for your help!

[yannp]

Hello,
It seems that dbt-unit-testing is always using uppercase identifiers with Snowflake which breaks with quoted identifiers containing lowercase characters. Could you confirm?

[dhwdeca]

Just ran into this issue as well.
It's not so much that dbt-unit-testing is using uppercase, but Snowflake will translate any column identifier that isn't quoted as an uppercase identifier.
eg. This query works in Snowflake:
"with test as (
select 'something' as "QUOTED_COLUMN")
select quoted_column from test"
This one doesn't and gives the same error we're running into:
"with test as (
select 'something' as "quoted_column")
select quoted_column from test"

[dcallananILMN]

I tested out the PR 194 changes as well and this looks like it will fix the issue.