Thanks for the bug report, Mathias.

I can confirm that -P and -i don't work well together.

I think correcting this bug is not too complex. It requires passing
the ignore-case options to PCRE while leaving the original pattern intact.
Would you be interested in submitting a patch?

Cheers,

TAA

-----------------------------------------------------
Tony Abou-Assaleh
Email:    [EMAIL PROTECTED]
Web site: http://tony.abou-assaleh.net
----------------------[THE END]----------------------

On Wed, 3 Oct 2007, Mathias wrote:

> Hey mighty GNU/grep people.
>
> When using perl-style, case-insensitive regular expressions, the input
> regex is lowercased. This also turns \W, \S and \D into \w, \s and \d,
> respectively. This can be worked around, but I still think it should
> be considered a bug, since there was no indications at first - all I
> got was empty stdout and stderr, and -c returned 0 lines.
>
> I used it to split input separated by whitespace - ^\s*\S+\s+\S+ etc.
>
> Thanks in advance, and thanks for a great to(ol|y).
>
> --
> Mvh. Mathias Rav
> Webudvikler hos Rav-stamps
> [EMAIL PROTECTED]
> --------
> Regards, Mathias Rav.
> Web developer at Rav-stamps
> [EMAIL PROTECTED]
>
>


Reply via email to