----- Original Message -----
From: Alfred Perlstein <bright@wintelcom.net>
To: Bruce Momjian <pgman@candle.pha.pa.us>
Cc: The Hermit Hacker <scrappy@hub.org>; Casey Lyon <casey@earthcars.com>; <pgsql-hackers@postgresql.org>
Sent: Sunday, April 29, 2001 11:17 PM
Subject: Re: [HACKERS] Thanks, naming conventions, and count()
> * Bruce Momjian <pgman@candle.pha.pa.us> [010429 20:14] wrote:
>
> > Yes, I like that idea, but the problem is that it is hard to update just
> > one table in the file. You sort of have to update the entire file each
> > time a table changes. That is why I liked symlinks because they are
> > per-table, but you are right that the symlink creation could fail
> > because the new table file was never created or something, leaving the
> > symlink pointing to nothing. Not sure how to address this. Is there a
> > way to update a flat file when a single table changes?
>
> Sort of, if that flat file is in the form of:
> 123456;"tablename "
> 000033;"another_table "
>
> ie, each line is a fixed length.
What if have one such a line in separate file in one dir?
Then there is no restriction on field length, you don't need
to dump the file each time and maintain the real .symlinks.
The 'lsdb' command (courtesy of The Hermit Hacker :))
will assemble all of them together and will show the DBA
where to look for a specific table.
File names can be your OIDs again, and just keep
table name inside the file. Keep these files under
a certain dir, and let the lsdb display them appropriately
when needed.
Or another idea is to create 'deferred' symlinks.
The (real) symlinks only created when DBA issues the 'lsdb'
command and lists them, and this list is maintained only
when the 'lsdb' is invoked....
Maybe this sounds stupid, but just a thought...
Serguei
