Re: InitPostgres and flatfiles question

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Bruce Momjian <bruce(at)momjian(dot)us>
Cc: Markus Schiltknecht <markus(at)bluegap(dot)ch>, pgsql-hackers(at)postgresql(dot)org
Subject: Re: InitPostgres and flatfiles question
Date: 2007-01-05 03:38:18
Message-ID: 9236.1167968298@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Bruce Momjian <bruce(at)momjian(dot)us> writes:
> Tom Lane wrote:
>> Markus Schiltknecht <markus(at)bluegap(dot)ch> writes:
>>> Is there a good reason to not let psql -c behave exactly like psql from
>>> STDIN?
>>
>> Backwards compatibility, mostly --- there seems to be a considerable
>> risk of subtly breaking people's scripts if we change the transactional
>> boundaries for psql -c commands.

> True, but if we keep hitting people who don't expect this behavior, I
> wonder if we should just fix it and mention it in the release notes.

One other point is that if we change -c's behavior, there won't be
*any* way to submit multiple queries in a single PQexec using plain
psql --- it will require hacking up a special test program using
libpq directly. Unless we have plans to obsolete
multi-queries-per-PQexec altogether, this doesn't seem like a good idea.

OTOH, you could argue that forbidding multiple queries in one PQexec
isn't a bad idea; it would provide an additional defense against
SQL-injection attacks. We did that already in the "extended" query
protocol and I've not heard many complaints.

I'd be willing to buy into doing both together, perhaps.

regards, tom lane

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Bruce Momjian 2007-01-05 03:43:22 Re: InitPostgres and flatfiles question
Previous Message Andrew Dunstan 2007-01-05 03:34:41 Re: ideas for auto-processing patches