| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | Pavel Golub <pavel(at)microolap(dot)com> |
| Cc: | pgsql-bugs(at)postgresql(dot)org |
| Subject: | Re: BUG #5451: pg_restore doesn't close input .tar archive |
| Date: | 2010-05-07 14:12:05 |
| Message-ID: | 11966.1273241525@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs |
Pavel Golub <pavel(at)microolap(dot)com> writes:
> TL> I'm not sure what distinction you're trying to draw. Do you want us to
> TL> add a close() just before exit()? If so, what for?
> First of all, for uniformity I guess. :) .backup files are closed
> properly.
> And secondly, I'm using these sources in my project and I
> need input file to be closed. Of course, I can find the place
> where to insert close() by myself, but I beleive that this must be done in the
> official release.
I don't think so. The pg_restore code is not written to be a library.
(Some have proposed making it into one, but there's a lot of work needed
to make that happen.) Even if it were designed as a library, I don't
think it would be the library's job to close the input file --- what
if the surrounding app passes it stdin, for instance?
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Bryan Henderson | 2010-05-08 23:42:12 | reference to undefined macro _MSC_VER |
| Previous Message | goutham jalakam | 2010-05-07 13:03:34 | Re: BUG #5450: system lacked sufficient buffer space, or because a queue was full, could not perform the operation |