From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | Lincoln Yeoh <lyeoh(at)pop(dot)jaring(dot)my> |
Cc: | Andrew Sullivan <andrew(at)libertyrms(dot)info>, pgsql-general(at)postgresql(dot)org |
Subject: | Re: The folding of unquoted names to lower case in |
Date: | 2003-03-15 16:04:44 |
Message-ID: | 28603.1047744284@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Lincoln Yeoh <lyeoh(at)pop(dot)jaring(dot)my> writes:
> Would a compile time setting be easier to do than a GUC? Everything
> unquoted folds up, and the system tables when doing initdb are all upper
> case (UGH!).
This might be doable as far as the backend itself goes, but I worry
about the repercussions for client-side code. To have a usable
installation you'd need libpq, psql, pg_dump, JDBC driver, etc etc
to all work in either mode. Not sure how far the implications extend
--- but our experience with the autocommit option leads me to be wary.
You could possibly get away with making pg_dump depend on the same
compile-time option as the backend, but for the client libraries in
general it would be a highly annoying restriction to have to be sure
all your machines are compiled the same way. libpq's
internally-generated queries are few and simple enough that we'd not
have much problem making sure they are case-agnostic, but can the same
be said of JDBC or ODBC?
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Aleksander Kmetec | 2003-03-15 18:16:19 | unicode sorting problems |
Previous Message | eivind.arnesen | 2003-03-15 10:25:29 | Point in time recovery? |