seer26 wrote: > ... > > So then you could have "non null terminated null terminated strings"? > Its just a trick to sneak embedded nulls into a utf-8 null terminated > string?...
Yes, if I decoded your meaning correctly. > if you want to have embedded nulls in a string, its probably better > not to use null-termination than to use overcoded utf-8. Except if you're using a language and libraries all built around null-termination (e.g., C). > This probably > represents some internal work-around for backwards compatibility, > and is not a real external encoding. It depends on how internal or external you mean. For JNI, I assume you'd consider it internal, because it's not stored anywhere except in memory. (And it's definitely for compatibility with C.) Java serialization is also for "externalizing" Java data (allowing it to be stored persistently outside of Java). Still, that's not usually intended as an interchange format, so that's probably not a "real external encoding." Daniel -- Daniel Barclay [EMAIL PROTECTED] -- Linux-UTF8: i18n of Linux on all levels Archive: http://mail.nl.linux.org/linux-utf8/
