Le 03/05/2017 à 19:29, Tom Lane a écrit :
> Adrian Klaver <adrian(dot)klaver(at)aklaver(dot)com> writes:
>> On 04/30/2017 11:54 PM, Philippe BEAUDOIN wrote:
>>> Just after the ALTER TABLE statement, I want to access this table. But
>>> at this time, the altered column is not visible with its new name.
>> From the error it looks to me like the statements are each run in a
>> separate session and the UPDATE is not seeing the ALTER TABLE.
> No, it's in the same session; the problem is the lack of a
> CommandCounterIncrement call between the ALTER's update and the parsing
> of the next statement. That means the update isn't visible yet,
> even in its own session. See the fix here:
>
> https://git.postgresql.org/gitweb/?p=postgresql.git;a=commitdiff;h=9209e07605afe0349660447f20d83ef165cdd0ae
>
> regards, tom lane
Thanks Tom for the fix. And thanks to Julien and Adrian too, for the
time spent on this issue.
Regards.