Re: Case expression pushdown - Mailing list pgsql-hackers

From Gilles Darold
Subject Re: Case expression pushdown
Date
Msg-id 5f5e4426-d01a-6dcb-4506-4cefe44b3301@migops.com
Whole thread Raw
In response to Re: Case expression pushdown  (Alexander Pyhalov <a.pyhalov@postgrespro.ru>)
List pgsql-hackers
Le 26/07/2021 à 18:03, Alexander Pyhalov a écrit :
> Tom Lane писал 2021-07-26 18:18:
>> Alexander Pyhalov <a.pyhalov@postgrespro.ru> writes:
>>> [ 0001-Allow-pushing-CASE-expression-to-foreign-server-v6.patch ]
>>
>> This doesn't compile cleanly:
>>
>> deparse.c: In function 'foreign_expr_walker.isra.4':
>> deparse.c:920:8: warning: 'collation' may be used uninitialized in
>> this function [-Wmaybe-uninitialized]
>>      if (collation != outer_cxt->collation)
>>         ^
>> deparse.c:914:3: warning: 'state' may be used uninitialized in this
>> function [-Wmaybe-uninitialized]
>>    switch (state)
>>    ^~~~~~
>>
>> These uninitialized variables very likely explain the fact that it fails
>> regression tests, both for me and for the cfbot.  Even if this 
>> weren't an
>> outright bug, we don't tolerate code that produces warnings on common
>> compilers.
>>
>>             regards, tom lane
>
> Hi.
>
> Of course, this is a patch issue. Don't understand how I overlooked this.
> Rebased on master and fixed it. Tests are passing here (but they also 
> passed for previous patch version).
>
> What exact tests are failing?
>

I confirm that there is no compilation warning and all regression tests 
pass successfully for the v7 patch, I have not checked previous patch 
but this one doesn't fail on cfbot too.


Best regards,

-- 
Gilles Darold




pgsql-hackers by date:

Previous
From: Robert Haas
Date:
Subject: Re: Some code cleanup for pgbench and pg_verifybackup
Next
From: Bharath Rupireddy
Date:
Subject: Re: CREATE SEQUENCE with RESTART option