On Thu, Mar 9, 2017 at 8:17 PM, Bruce Momjian <bruce@momjian.us> wrote:
>> In my opinion, we expose query id (and dbid, and userid) as the
>> canonical identifier for each pg_stat_statements entry, and have done
>> so for some time. That's the stable API -- not query text. I'm aware
>> of cases where query text was used as an identifier, but that ended up
>> being hashed anyway.
>
> Speaking of hash values for queries, someone once asked me if we could
> display a hash value for queries displayed in pg_stat_activity and
> pg_stat_statements so they could take a running query and look in
> pg_stat_statements to see how long is usually ran. It seemed like a
> useful idea to me.
I agree.
> I don't think they can hash the query manually because of the constants
> involved.
It would be a matter of having postgres expose Query.queryId (or the
similar field in PlannedStmt, I suppose). Morally, that field belongs
to pg_stat_statements, since it was written to make the query
normalization stuff work, and because everything would break if
another extension attempted to use it as pg_stat_statements does.
Whether or not that makes it okay to expose the hash value in
pg_stat_activity like that is above my pay grade, as Tom would say.
--
Peter Geoghegan