grep --ignore-case efficiency

[Paulo Nogueira]

 Hello,

 this is not about a bug in the usual sense, rather
 about the poor(?) efficiency of "grep --ignore-case"
 as compared to the default do-not-ignore-case.

 Although I am completely oblivious of the inner workings
 of grep, it seems rather strange that the ignore-case
 could take (say) tens of times longer to execute than
 the do-not-ignore-case -- and I am referring to normal
 ASCII files.

 Provided there is an efficient way to convert from one
 character case to the other, as there is in ASCII (say)
 it seems pretty obvious that one could achieve a much
 smaller execution-time ratio. Is the need to accommodate
 for a generic character set the problem? What is?

 Best regards
  Paulo