Library function to detect UTF-8 streams without BOM

I'm looking for a function/algorithm to detect UTF-8 files without the byte
order mark. There's some detection code in TNT, for example (the last free
version), but it relies solely on the BOM. I've tried jclUnicode and the
DIUnicode package (just the help file for the latter), but no go. Perhaps
the new version of TNT from TMS has something?

Thanks,
marek



--
No ads, no nags freeware: http://www.tranglos.com


.