Re: pgsql/src/bin/pg_dump pg_dumpall.sh - Mailing list pgsql-committers

From Bruce Momjian
Subject Re: pgsql/src/bin/pg_dump pg_dumpall.sh
Date
Msg-id 200204112103.g3BL31A03332@candle.pha.pa.us
Whole thread Raw
In response to Re: pgsql/src/bin/pg_dump pg_dumpall.sh  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: pgsql/src/bin/pg_dump pg_dumpall.sh
List pgsql-committers
Tom Lane wrote:
> momjian@postgresql.org (Bruce Momjian - CVS) writes:
> > Modified files:
> >     src/bin/pg_dump: pg_dumpall.sh
>
> > Log message:
> >     Fix pg_upgrade to handle dbnames, user/group names with spaces.
>
> I think it's illusory to imagine that this fixes the issue.  Unless
> "read"'s behavior is more tightly specified than I think, you are still
> gonna have trouble with leading, trailing, or consecutive blanks.

Yes, it does fail on leading/trailing spaces, but it is better than it
was before.

> I'm also concerned that you've opened up portability issues concerning
> newlines embedded in shell literals (eg, does that sed command portably
> do what you think?).

I certainly think so.  It is:

    echo "a b c" | sed 's/ /\
/g'

outputs:

    a
    b
    c

Embedding backslash-n in 'sed' is a portability problem.

Clearly read has problems:

    #$ read X
       a   b
    #$ echo $X
    a b

I could so IFS="newline" to handle the space issue.  Let me try that.

--
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 853-3000
  +  If your life is a hard drive,     |  830 Blythe Avenue
  +  Christ can be your backup.        |  Drexel Hill, Pennsylvania 19026

pgsql-committers by date:

Previous
From: Tom Lane
Date:
Subject: Re: pgsql/src/bin/pg_dump pg_dumpall.sh
Next
From: Bruce Momjian
Date:
Subject: Re: pgsql/src/bin/pg_dump pg_dumpall.sh