Re: OK, that's one LOCALE bug report too many... - Mailing list pgsql-hackers

From teg@redhat.com (Trond Eivind GlomsrØd)
Subject Re: OK, that's one LOCALE bug report too many...
Date
Msg-id 20001126211200.087342801F@halden.devel.redhat.com
Whole thread Raw
In response to OK, that's one LOCALE bug report too many...  (Tom Lane <tgl@sss.pgh.pa.us>)
Responses Re: OK, that's one LOCALE bug report too many...
List pgsql-hackers
Tom Lane <tgl@sss.pgh.pa.us> writes:

> Also, since "LC_COLLATE=en_US" seems to misbehave rather spectacularly
> on recent RedHat releases, I propose that initdb change "en_US" to "C"
> if it finds that setting.  

It does not misbehave in glibc (it's not Red Hat specific).
Basically, glibc is the old


From a discussion on a semi-internal list, written by Alan Cox:

************************************************************************
I read the ISO doc (god Its boring)

Ok

Ulrich is right for the spec. Its the official correct filing order
for more
than just in computing

I think the right answer maybe this

Default to ISOblah including sort remaining sorting AbBb..
Document this and also how to switch just the collation series to Unix
style
in the README files and docs that come with the release (like we
documented
how to turn off color ls


Ultimately this comes down to:        
Unix behaviour since 197x versus librarians and others since
considerably
earlier. We are breaking Unix behaviour but I can now sort of
appreciate
the thinking behind this. 

************************************************************************


-- 
Trond Eivind Glomsrød
Red Hat, Inc.


pgsql-hackers by date:

Previous
From: Bruce Momjian
Date:
Subject: Re: LIKE optimization and locale
Next
From: teg@redhat.com (Trond Eivind GlomsrØd)
Date:
Subject: Re: OK, that's one LOCALE bug report too many...