Re: 7.4.1 release status - Turkish Locale - Mailing list pgsql-hackers

From Tom Lane
Subject Re: 7.4.1 release status - Turkish Locale
Date
Msg-id 24974.1076364601@sss.pgh.pa.us
Whole thread Raw
In response to Re: 7.4.1 release status - Turkish Locale  (Greg Stark <gsstark@mit.edu>)
Responses Re: 7.4.1 release status - Turkish Locale  ("Nicolai Tufar" <ntufar@pisem.net>)
List pgsql-hackers
Greg Stark <gsstark@mit.edu> writes:
> But the example given was "SERIAL". "serial" is an English word, not a
> Turkish word. It shouldn't really be subject to Turkish locale effects
> at all.

SERIAL is not a keyword according to the grammar.  Neither are PUBLIC,
VOID, INT4, and numerous other examples.  It's not appropriate to try to
fix this by making them all keywords --- that will just create other
problems.  (And where do you draw the line, anyway?  Should every
identifier present in the default system catalogs become a keyword?)

> I'm wondering if he really expects all identifiers to be subject to
> this ascii downcasing.

Without doubt it isn't ideal, but if we don't do something then a lot of
stuff starting with initdb is broken.  We could perhaps work around the
problem by spelling everything in lower-case in all the commands we
issue, but I can't see that as an acceptable answer either.  We can't
expect to control all the SQL sent to a database.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Greg Stark
Date:
Subject: Re: 7.4.1 release status - Turkish Locale
Next
From: Josh Berkus
Date:
Subject: Re: Proposed Query Planner TODO items