Re: Support prepared statement invalidation when result types change

From: Andy Fan <zhihui(dot)fan1213(at)gmail(dot)com>
To: Jelte Fennema <me(at)jeltef(dot)nl>
Cc: PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Support prepared statement invalidation when result types change
Date: 2023-09-14 23:41:46
Message-ID: CAKU4AWpd82LBE=T-=Dyf_6cHSP+sbo9-tKH5tdxEu5V0BX0yUA@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi,

>
> This requirement was not documented anywhere and it
> can thus be a surprising error to hit. But it's actually not needed for
> this to be an error, as long as we send the correct RowDescription there
> does not have to be a problem for clients when the result types or
> column counts change.
>

What if a client has *cached* an old version of RowDescription
and the server changed it to something new and sent resultdata
with the new RowDescription. Will the client still be able to work
expectly?

I don't hope my concern is right since I didn't go through any of
the drivers in detail, but I hope my concern is expressed correctly.

--
Best Regards
Andy Fan

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Andres Freund 2023-09-15 00:33:35 Re: Have better wording for snapshot file reading failure
Previous Message Michael Paquier 2023-09-14 23:28:26 Re: [PATCH] Add inline comments to the pg_hba_file_rules view