Hello Tom,
Thanks for taking the time.
> Is that theexact query you're issuing?
Yes, that's the exact query.
You can reproduce the issue by creating the table and running the C program from my previous e-mail.
‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐
On Thursday, 1 April 2021 11:14, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Giovani Garcia giovani.garcia@protonmail.com writes:
>
> > The issue I'm finding is that the Oid returned for a VARCHAR column is TEXTOID (25) instead of VARCHAROID (1043).
>
> This isn't hugely surprising in general. varchar has no operations
> of its own; PG treats it as sort of a domain over text. So if you
> do anything at all to a varchar column, it'd be quite likely for
> the column to be implicitly coerced to text within the query.
>
> > Now, when I run the following program (simplified for brevity),
>
> > "SELECT key FROM oid_test WHERE value = $1",
>
> Is that theexact query you're issuing? I don't see anything
> in that that would cause a coercion, but maybe you left out some
> relevant detail.
>
> regards, tom lane