On Tue, 2006-04-11 at 15:47 +0200, Tomas Larsson wrote: > Yes, the chars are placed at the same place, but UTF8 uses two bytes > to represent the char, when ISO8859 uses one byte, for the lower 256 > bytes. I finally got my hands on a document explaining this, it might have helped if I could have got the Unicode site to load yesterday, but the network said "no". :-\ The Wikipedia page I referred to that did load was a little scant in precise descriptions. In ASCII, characters 0 to 127; and in ISO-8859-1, characters 0 to 127; and in UTF-8, characters 0 to 127, are all the same characters, and are all represented by the same bytecodes. e.g. The letter A is character 65 in all of them. But once you get above 127, they differ: ASCII *ONLY* covers 0 to 127. If it has characters higher than 127, it's not "ASCII". ISO-8859-1 encodes characters 0 to 255 in eight bits directly (character number 214 is represented by the value 214). UTF-8 starts using its special multi-byte scheme from 128 upwards, below that it refers to them directly (e.g. character 90 is represented by the value 90). (This differs with what I've read before.) While the 214th character is the same character in both ISO-8859-1 and UTF-8, the *code* used to represent it is different in both (it's "214" in ISO-8859-1, but the codes 0xC3 0x96 in conjunction for UTF-8). > Since MYSQL is using the system settings, when inputting a Ö it was > transformed to the 2-byte UTF8 format. One would hope that you could specify what system for the database to use, regardless of the host OS, as a database may be used between numerous different computers. Presuming they all ran the same would lead to errors. -- (Currently running FC4, occasionally trying FC5.) Don't send private replies to my address, the mailbox is ignored. I read messages from the public lists.
