Tsql not equal workaround

[dataders]

resolves #3001
@jtcohen6 @mikaelene

Description

per this conversation
microsoft/dbt-synapse#32 (comment)

Checklist

• I have signed the CLA
• I have run this code in development and it appears to resolve the stated issue
• This PR includes tests, or tests are not required/relevant for this PR
• I have updated the CHANGELOG.md and added information about my change to the "dbt next" section.

[cla-bot]

Thanks for your pull request, and welcome to our community! We require contributors to sign our Contributor License Agreement and we don't seem to have your signature on file. Check out this article for more information on why we have a CLA.

In order for us to review and merge your code, please submit the Individual Contributor License Agreement form attached above above. If you have questions about the CLA, or if you believe you've received this message in error, don't hesitate to ping @drewbanin.

CLA has not been signed by users: @swanderz

[dataders]

strange redshift error.. perhaps it is transient?

=================================== FAILURES ===================================
_ IncrementalTestConcurrentTransaction.test__redshift__concurrent_transaction_incremental _
[gw3] linux -- Python 3.6.9 /home/dbt_test_user/project/.tox/integration-redshift-py36/bin/python

self = <test_concurrent_transaction.IncrementalTestConcurrentTransaction testMethod=test__redshift__concurrent_transaction_incremental>

    @use_profile("redshift")
    def test__redshift__concurrent_transaction_incremental(self):
        self.reset()
>       self.run_test()

test/integration/032_concurrent_transaction_test/test_concurrent_transaction.py:136: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ 
test/integration/032_concurrent_transaction_test/test_concurrent_transaction.py:102: in run_test
    self.assertEqual(self.query_state['view_model'], 'good')
E   AssertionError: 'error: UDF working memory threshold excee[280 chars]\n\n' != 'good'
E   + good- error: UDF working memory threshold exceeded
E   - DETAIL:  
E   -   -----------------------------------------------
E   -   error:  UDF working memory threshold exceeded
E   -   code:      10002
E   -   context:   
E   -   query:     0
E   -   location:  stream.cpp:41
E   -   process:   padbmaster [pid=2255]
E   -   -----------------------------------------------
E   -

[dataders]

also git newb... but I made my two commits into a new branch directly after running git reset --hard upstream/dev/kiyoshi-kuromiya, so i'm confused why those 6 commits from two years ago are showing up...

[jtcohen6]

Nice! Ok, I'm up for sneaking this into v0.19.0, since we already have some eyes on snapshots (on account of invalidate_hard_deletes).

Copying from microsoft/dbt-synapse#32:

select

  scenario,
  (a is null) != (b is null) as previous_logic,
  ((a is null) and not (b is null)) or (not (a is null) and (b is null)) as proposed_logic

from (

  select 1 as scenario, null as a, null as b
  union all
  select 2 as scenario, 1 as a, 1 as b
  union all
  select 3 as scenario, null as a, 1 as b
  union all
  select 4 as scenario, 1 as a, null as b

) sbq

scenario	previous_logic	proposed_logic
1	false	false
2	false	false
3	true	true
4	true	true

This works on every database I can think to test: Postgres, Redshift, Snowflake, BigQuery, Spark, Presto.

Looks like the Redshift error was intermittent—sometimes the UDF memory is exhausted, we're not exactly sure why—so I reran the tests with success.

Tiny tiny requested changes re: changelog note + comments. As far as the extraneous git commit, I think those are hanging out on your fork's version of dev/kiyoshi-kuromiya (see diff). You could try:

• Deleting your fork's branch and recreating from remote
• Rebasing your branch against fishtown-analytics/dbt/dev/kiyoshi-kuromiya

[mikaelene]

This is great! Will make dbt-sqlservers life much easier