On Wed, Nov 01, 2006 at 08:50:30PM +0100, Martijn van Oosterhout wrote: > > Could this paragraph be put into the docs and/or the FAQ, > > please ? Along with the recommendation that if you require > > multiple encodings for your databases you better had your OS > > locale configured properly for UTF8 and use UNICODE > > databases or do initdb with the C-locale. > > Err, multiple encodings don't work full-stop. Well, yes, I was thinking of multiple client encodings which can be supported either via a C-locale-initdb with the databases set to the encoding you require (but sorting/etc won't work, I know) or by doing a unicode-initdb and using unicode databases. In each case the client encodings can be "multiple" ones - as long as conversion is possible. Sorting etc may still be wrong, but at least the proper characters are going in and coming back. > Any particular locale (as > defined by POSIX) is only really designed to work with one encoding. Sure. What I meant is that if you have a unicode database you can use several client encodings and get back the properly encoded characters. > The fact that the C locale produces an order when sorting UTF8 text is > really just luck. Yes. > > Here are a few data points from my Debian/Testing system in > > favour of not worrying too much about installed ICU size as > > it is being used by other packages anyways: > > We'd need a suitable patch first before we start worrying about that. I > think diskspace is less of an issue now. Well, size did come up in a "recent" discussion so I thought I'd mention the above facts. Karsten -- GPG key ID E4071346 @ wwwkeys.pgp.net E167 67FD A291 2BEA 73BD 4537 78B9 A9F9 E407 1346
