http://qs321.pair.com?node_id=1011959


in reply to Issue with reading a unicode file

As a guess, since you seem to be on Windows, your input file is likely to begin with a Byte Order Mark (BOM), which Microsoft uses as a convention to distinguish the various flavors of UTF. A UTF-8 byte order mark would be three bytes long,  0xEF,0xBB,0xBF. In perl, it appears as the code point  "\N{U+FEFF}". You could try  tr/\N{U+FEFF}//d to remove it.