Hi Dave,

Thanks for the response. I think you are definitely right about the
sorting order being correct (a feature and not a bug), but I think that
the much slower speed is a bug. It might be a sort of wishlist type bug,
but there's no theoretical reason why a faster method of sorting can't
be used (even though there might be a practical reason not to: no demand
from users or no time for developers). It may be that the algorithm used
for sorting 8-bit ASCII will not perform well for 16-bit random
collating order.

But of course the locale specific unicode can be converted to a
representation that sorts quickly. For example the first character in
collating order could become 0, the second becomes 1, and so on to
2^16-1 for example (for 16-bit characters). Then you can sort it using
plain ASCII sorting rules.

Bijan

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/846628

Title:
  gnu sort extremely slow in non C locale

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/coreutils/+bug/846628/+subscriptions

-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to