|
XML allows "native language markup", which is where element and
attribute names can by and large use the typical graphical characters
used for words in any native language. So Chinese can have element
names entirely with ideographic characters and so on. (The only proviso
is that spaces and apostrophe's are not possible.) This idea was adopted into XML following the ERCS principles of the CJK DOCP group, an ISO-liaison expert group made up from standards people, industry and academics from East Asia, in the mid 1990s (I wrote it.) After XML, the principles have been consolidated by W3C and Unicode in a joint technical report concerning characters suitable for use in markup. So Turkish dotless i is certainly allowed as an XML name character. (In fact, it is also the main reason why XML is case-sensitive, IIRC: it means there is no nation-neutral case-mapping strategy for A-Z.) One possible reason there may a complaint about that character is if you are using the wrong encoding declaration. Your document should be using UTF-8, or 8859-9 (or 8859-3, or CP1254 etc). Many character sets do not have enough redundant code-points to allow incorrect labeling to be determined (for example, between the 8859-n character sets). So the strict naming rules of XML 1.0 serve as a back up to detect when code comes through that is not allowed as part of a name: it is a sign that there has been a bug or data corruption and prevents further infection. When looking at character encoding, the golden rule is USE A HEX EDITOR. Don't open a file in some vanilla text editor unless you are really clear what encodings it reads, how it handles fonts, and what input mappings it may perform. Cheers Rick Jelliffe Desmond Kirrane wrote: 5585ca8d0712130334k111cb306wa5f13b91c3518c70@mail.gmail.com" type="cite">Hi, |