Comment on \d less efficient than [0-9]Comments−foobar__13yThe fact that character ranges like [a-z] can depend on the value of LC_COLLATE is also something not many people are aware of. $ echo "ä" | LC_COLLATE=C grep '[a-z]' $ echo "ä" | LC_COLLATE=en_US.UTF-8 grep '[a-z]' ä For common values of LC_COLLATE, the range [a-z] does not exclude accented characters and umlauts.
Comments
The fact that character ranges like [a-z] can depend on the value of LC_COLLATE is also something not many people are aware of.
For common values of LC_COLLATE, the range [a-z] does not exclude accented characters and umlauts.