On Mon, 22 Aug 2011 19:22:09 +0200 Colomban Wendling lists.ban@herbesfolles.org wrote:
grep uses plain char and doesn't support UTF-8. [...]
...but б@боза@@ will match (at least using UTF-8), [...]
Neither the character classes nor the upper/lower cases will work properly if you feed an 8-bit char program with UTF-8. grep -i -w produce lots of false positives and/or negatives, and so will the [] sets. Except for the 8859-1 characters, of course.