WIP: first commit - added aliasing to unique and not_null

[tnightengale]

resolves #2061

Description

Small fix for the bug identified in this issue: #2061.

Added aliasing to the unique and not_null schema tests to fix bug where they would fail if model and column had the same name.

Changes

core/dbt/include/global_project/macros/schema_tests/not_null.sql -> added alias
core/dbt/include/global_project/macros/schema_tests/unique.sql -> added alias

Tests

Hoping that someone from fishtown can kick of the CI tests - I did not run the make file tests locally. I think the aliasing is completely standard sql and should be db agnostic - I am slightly concerned if aliased.{{ column_name }} will compile with a space to aliased. some_column.

[cla-bot]

Thank you for your pull request and welcome to our community. We could not parse the GitHub identity of the following contributors: Teghan Nightengale.
This is most likely caused by a git client misconfiguration; please make sure to:

1. check if your git client is configured with an email to sign commits git config --list | grep email
2. If not, set it up using git config --global user.email email@example.com
3. Make sure that the git commit email is configured in your GitHub account settings, see https://github.com/settings/emails

[drewbanin]

Hey @tnightengale - I just took a quick look here and dropped a comment in below. Let me know what you think about that.

Also, you'll need to sign the CLA signed to get this PR merged. Please give me a shout whenever you're able to do that. Thanks!

[drewbanin]

I just thought of something really critical here - you can currently supply an expression to these tests, eg:

models:
 - name: my_model
    tests:
      - unique:
          arg: concat(field_1, field_2)

I think code like this will no longer work if we purse this approach :/

Maybe an alternative idea is to do something like:

select count(*)
from (
  select {{ column_name }} as __dbt_not_null_field from {{ model }}
) as __dbt_test_sbq
where __dbt_not_null_field is null

I think this should support existing use cases while also addressing the current issue with BigQuery table names masking column names.

[drewbanin]

see above

[drewbanin]

hey @tnightengale - going to close this one out as stale for now, but please let me know if you're interested in picking this one back up!