From: | "Thomas G(dot) Lockhart" <lockhart(at)alumni(dot)caltech(dot)edu> |
---|---|
To: | Patrick Welche <prlw1(at)cam(dot)ac(dot)uk> |
Cc: | hackers(at)postgreSQL(dot)org |
Subject: | Re: [HACKERS] need help with csh |
Date: | 1998-10-06 01:34:07 |
Message-ID: | 3619738F.51DA1360@alumni.caltech.edu |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
Patrick Welche wrote:
>
> > And how shall it dsitinguish between a path and a variable?
> > > setenv PGDATA2 /home/postgres/data
> > > initlocation PGDATA2
> > And how shall it distinguish between a path and a variable?
> > But this could mean the path PGDATA2 too, couldn't it?
I test that the argument, when translated, has a non-empty result. If it
doesn't, I assume that it is an actual path.
> initlocation $PGDATA2
Yes, that's how it has always worked. I wanted it to also function
correctly when invoked with the environment variable as an argument (if
someone forgets to put in the "dollar sign").
I have a working version committed to the cvs tree which uses printenv
to extract the contents of the argument. It correctly distinguishes
between paths and envars, and does the right thing with both.
I'd appreciate any reports of problems on any platforms or environments.
Thanks all for the tips.
- Tom
From | Date | Subject | |
---|---|---|---|
Next Message | Bruce Momjian | 1998-10-06 01:56:50 | Re: [HACKERS] Open 6.4 items |
Previous Message | Tom Lane | 1998-10-05 23:39:39 | Re: [HACKERS] TCL_ARRAYS code in libpgtcl is pretty seriously broken |