Re: [HACKERS] postgres init script things solved - Mailing list pgsql-hackers

From Peter Bentley
Subject Re: [HACKERS] postgres init script things solved
Date
Msg-id 3545A079.226C749D@nomura.co.uk
Whole thread Raw
List pgsql-hackers
> > >OK, but _I_ don't run bash. So someone else is now maintaining this
> > >file? Why didn't we keep both forms in the file, with one commented out?
> > >What are we trying to accomplish here??
> ...
> > About bash, it's the usual shell for regular users. I guess the start/stop
> > script was meant for users who don't know/cannot write their own.
>
> Perhaps we should write scripts for just plain old 'sh'. This is available
> everywhere (sort of even on SCO).
>
> If not that, bash might be a better choice than tcsh as it is perhaps more
> common (we even run it on all our Solaris machines where I work).

If you want portability, then plain old Bourne shell should probably be
the shell of choice.  Most implementations's on /bin/sh behave the same,
and such scripts will run just fine under bash or ksh.

The same is not true in the csh world...different implementations of
/bin/csh tend to have different bugs (eg some of them have the sense of
|| and && swapped).  Tcsh is (I believe) fairly bug free...but much less
widely available than /bin/sh...

Pete.

pgsql-hackers by date:

Previous
From: Hannu Krosing
Date:
Subject: Re: New Driver and Unique Indexes
Next
From: Ralf Berger
Date:
Subject: Re: [QUESTIONS] Postgres still dying on insert