[bug-grep] Re: grep: -i option not working i cronjobs

[Julian Foad]

Tony Abou-Assaleh wrote:
> 1) Have 2 buffers, original and TOLOWERCASE'ed, and when printing the
> output, print it from the original buffer
> 2) convert the regular expression to match both cases for any literals
> simply by expanding each alphabetic symbol to a character class of it's
> lowercase and upper case forms: e.g, 'a' -> '[aA]', 'B'-> '[bB]', etc.
>
> My guess is that for most cases, the second solution would be faster. But
> that's just a semi-educated guess at the moment. Does anyone know better?

I agree that (2) is the better solution.  It better matches the intended 
meaning, and will work properly in situations like languages that have no 
lower-case equivalent of a certain character, or have more than one lower-case 
version of a single upper-case character.

- Julian