Re: when options conflict

[Tony Abou-Assaleh]

On Sun, 21 May 2006, Bob Proulx wrote:
> Tony Abou-Assaleh wrote:
> > > >    -- Aborting in error for -G/-F/-E/-P,
> > >
> > > Definitely.  Otherwise someone might think that something
> > > like 'grep -E -e extendedpattern -P -e perlpattern' would work.
> >
> > It would be awesome if one could do that! This should go into feature
> > requests.
>
> How often would syntax like that really be used?  Sure it would be
> cool.  But I think people wanting pcre patterns would simply want them
> all of the time.  Is there really enough need to match different
> styles of regular expressions to warrent the work to put that into the
> code?  I doubt it would be worth it in actual usage.

For most everyday usages by most users, the current feature set is
sufficient. I can surely think of situations where the above syntax may
make a lot of sense.

Example: you have complex patterns, in general -E is faster than -P, but
some patterns require the features of -P. If you have a large file, and
these days it is not uncommon to have large files (or a lot of small
ones), such optimization may give noticeable difference in performance.

I haven't done benchmarking, so the above is hypothetical.

Since we're not doing any new features at the moment, nor in the near
future, and such a feature as above would be of somewhat low priority due
to its complexity, my comment was half sarcastic.

Cheers,

TAA

-----------------------------------------------------
Tony Abou-Assaleh
Lecturer, Computer Science Department
Brock University, St. Catharines, ON, Canada, L2S 3A1
Office: MC J215
Tel:    +1(905)688-5550 ext. 5243
Fax:    +1(905)688-3255
Email:  address@hidden
WWW:    http://www.cosc.brocku.ca/~taa/
----------------------[THE END]----------------------