On 4/2/19 8:00 AM, Rich Shepard wrote:
> An unexpected result this morning that I'd like to understand. In emacs I
> loaded the file organizations.sql in a buffer to insert new rows in that
> table. Rather than running psql from a virtual console, I opened an emacs
> shell and entered the command,
> psql -f organizations.sql -d bustrac
> and saw an error that the last line in the .sql file was in conflict
> with an
> existing primary key. (I let postgres assign the keys when new rows are
> inserted.)
What was the exact message?
What was in the last line?
>
> Checking the table for that primary key confirmed that it was a different
> organization so there should not have been a conflict.
>
> Using a virtual console, and in the directory with the scripts, I ran the
> same psql command and it completed as expected.
>
> I don't think this is a emacs error because the shell opens in the same
> directory as the script I'm running.
>
> Why might postgres fail to run the script within an emacs shell while
> having
> no issues running the same script from a virtual console?
Difference in shell escaping?
>
> A curious mind wants to learn.
>
> TIA,
>
> Rich
>
>
>
--
Adrian Klaver
adrian.klaver@aklaver.com