>> Meanwhile, am I correct in assuming that re-initialising my database cluster >> with "--locale=C" will solve the problem? > > AFAIK it should --- of course you won't get any very intelligent sorting > or case folding, but at least it can tell the difference between > different characters ;-). Be sure to still use encoding = unicode. Okay, thanks for confirming that (!) But... will setting the C locale affect how PostgreSQL sorts Latin-1 words? If it does, I'm in trouble! I need to be able to sort (using ORDER BY queries) English and other European names/words. (I've tried searching for this topic, but failed to find an answer.) Would specifying a locale for LC_COLLATE take care of this? Perhaps something like this: initdb --locale=C --lc-collate=en_GB.UTF-8 --encoding UNICODE I'm sorry to drag this out. Perhaps I cannot have it both ways. Kind regards Harry Mantheakis London, UK ---------------------------(end of broadcast)--------------------------- TIP 5: don't forget to increase your free space map settings
