X-Original-To: pgsql-admin-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id C163955791 for ; Thu, 24 Feb 2005 17:22:52 +0000 (GMT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 21868-04 for ; Thu, 24 Feb 2005 17:22:44 +0000 (GMT) Received: from ns1.turtle-entertainment.de (ns1.turtle-entertainment.de [193.41.200.20]) by svr1.postgresql.org (Postfix) with ESMTP id 0E0E15574D for ; Thu, 24 Feb 2005 17:22:42 +0000 (GMT) Received: from ip96.85.1311c-cud12k-02.ish.de ([62.143.85.96] helo=mail.office.turtle-entertainment.de) by ns1.turtle-entertainment.de with asmtp (Exim 3.22 #3 (Debian)) id 1D4MhO-0004bS-00; Thu, 24 Feb 2005 18:22:50 +0100 Received: from bm.office.turtle-entertainment.de ([212.6.194.203]) by mail.office.turtle-entertainment.de with asmtp (Exim 3.22 #7 (Debian)) id 1D4MhO-0004Ul-00; Thu, 24 Feb 2005 18:22:50 +0100 Message-ID: <421E0D6B.6010303@turtle-entertainment.de> Date: Thu, 24 Feb 2005 18:22:51 +0100 From: Bjoern Metzdorf User-Agent: Mozilla Thunderbird 1.0 (Windows/20041206) X-Accept-Language: en-us, en MIME-Version: 1.0 To: Tom Lane Cc: "Pgsql-Admin (E-mail)" Subject: Re: invalid multibyte character for locale References: <421CF15C.6010306@turtle-entertainment.de> <421CF1CE.3000800@turtle-entertainment.de> <23514.1109203127@sss.pgh.pa.us> <421D8541.2000904@turtle-entertainment.de> <24832.1109258150@sss.pgh.pa.us> In-Reply-To: <24832.1109258150@sss.pgh.pa.us> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=1.842 tagged_above=0 required=5 tests=FORGED_RCVD_HELO, RCVD_IN_NJABL_DUL, RCVD_IN_SORBS_DUL X-Spam-Level: * X-Archive-Number: 200502/287 X-Sequence-Number: 16455 Tom Lane wrote: > I don't know what behavior you thought you were getting from upper/lower > on UTF-8 data in 7.4, but it was surely not correct. If you want to > duplicate that misbehavior, try SQL_ASCII with C locale. This does not > stop you from storing UTF-8 in your database, mind you --- it just > loses validation of encoding sequences and conversion to other schemes. > But having said that, upper() should work if the locale matches the > encoding. You might take the trouble to trace down exactly what data > value it's barfing on. I want to keep UNICODE encoding in any case. So you say, that 7.x just did not cope at all with multibyte chars and upper() and lower() spit out what the C functions toupper and tolower spit out? I also want to stay with locale C, because of the speed. I have different languages, not only one specific, so changing the locale would not help at all. I assume I could just remove #define USE_WIDE_UPPER_LOWER from oracle_compat.c to emulate the old behaviour. But a cleaner fix would be to check if we are using UNICODE and locale is C or POSIX and only then skip USE_WIDE_UPPER_LOWER. Comments? Regards, Bjoern