Re: Alternative to psql -c ? - Mailing list pgsql-general

From Merlin Moncure
Subject Re: Alternative to psql -c ?
Date
Msg-id CAHyXU0xrhGc+qsmAT1mZ3SKwqGh0SuKSTfWMp-cqnt-YVtp_rQ@mail.gmail.com
Whole thread Raw
In response to Re: Alternative to psql -c ?  (James Le Cuirot <chewi@aura-online.co.uk>)
List pgsql-general
On Thu, Jun 26, 2014 at 7:59 AM, James Le Cuirot
<chewi@aura-online.co.uk> wrote:
> On Thu, 26 Jun 2014 07:23:02 -0500
> Merlin Moncure <mmoncure@gmail.com> wrote:
>> To be clear, Tom was advising not to rely on some of the quirky
>> aspects of -c.  psql as it stands right now has a some limitations:
>> single transaction mode does not work with stdin and there is no
>> reasonable way to pass multiple scripts through the command line.
>> Adding it up this means that for generic multiple .sql passing you
>> have to wrap psql with a script.
>
> I never said that I wanted to pass multiple scripts, just singular
> scripts who's contents might be too large for the command line limit,
> and that they might contain transaction statements.

yup.  Passing huge scripts via -c is definitely not a good practice.
-c is designed for things like firing off a stored procedure from cron
or bash variable assignment (for example: FOO=`psql -tXAqc"select
bar()"`)

> These scripts come from a Ruby string buffer so passing them via stdin
> would be preferable to having to dump them out to a file first. You say
> that single transaction mode doesn't work with stdin but it looks like
> this was fixed in 9.3.

yup -- i keep forgetting that.  too many years of not having it i suppose.

merlin


pgsql-general by date:

Previous
From: Adrian Klaver
Date:
Subject: Re: python modul pre-import to avoid importing each time
Next
From: Shaun Thomas
Date:
Subject: Re: Fast data, slow data