From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Gregory Stark <stark(at)enterprisedb(dot)com> |
Cc: | "daveg" <daveg(at)sonic(dot)net>, "Andrew Dunstan" <andrew(at)dunslane(dot)net>, "Davy Durham" <pubaddr5(at)davyandbeth(dot)com>, pgsql-patches(at)postgresql(dot)org |
Subject: | Re: Feature: give pg_dump a WHERE clause expression |
Date: | 2008-06-02 14:17:02 |
Message-ID: | 18023.1212416222@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-patches |
Gregory Stark <stark(at)enterprisedb(dot)com> writes:
> "daveg" <daveg(at)sonic(dot)net> writes:
>> The argument that one could just use copy and tar and gzip applies to the
>> whole existance of pg_dump itself.
> That's not really true, pg_dump does a *lot* more work than just group table
> data together. Its real value comes in reconstructing DDL for all the objects
> and understanding the dependencies between them. That's functionality that
> isn't available elsewhere and can't be reproduced without just reimplementing
> pg_dump.
Indeed. The other altar that pg_dump spends a lot of time worshipping
at is cross-version compatibility. It's got quite enough on its plate
without people trying to convert it into an ETL tool.
Which is not to say that we don't need an ETL tool, just that it should
be separate from pg_dump.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Tom Lane | 2008-06-02 14:37:29 | Re: extend VacAttrStats to allow stavalues of different types |
Previous Message | Heikki Linnakangas | 2008-06-02 13:12:03 | Re: Map forks (WIP) |