> I don't know UTF-8 very well. I am of the impression that > 0x00 can occur > in UTF-8, in multi-byte character sequences. You've been > researching the > UTF-8. Can you determine if that's true? If it is, then we > cannot limit > octets to 1..255 values.
It can not happen. The reason is that all octets 0..127 are reserved for US-ASCII, so these are never reused in multibyte sequences. These octets are all in the range 128..255. Anton Okmianski has independently verified this (this message has not been propagated to all list subscribers). Of course it can happen in deliberately invalidly crafted messages, just as a reminer Rainer
