RE: Pattern Match within PARI/GP

["Kerl, John" <John.Kerl@Avnet.com>]

For what it's worth, this is precisely the sort of thing
Perl excels at.  E.g. gp myfile.gp > out.txt; perl myperl.pl out.txt
where myperl.pl does something like
$line=$_; print $line if $line =~ m/xyxyxy/;
Remember that Perl handles those massive datasets in the Human
Genome Project.

-----Original Message-----
From: Bill Allombert [mailto:allomber@math.u-bordeaux.fr]
Sent: Wednesday, February 18, 2004 4:28 PM
To: pari-users@list.cr.yp.to
Subject: Re: Pattern Match within PARI/GP


On Wed, Feb 18, 2004 at 12:18:21PM -0800, Paulo Ney de Souza wrote:
> 
> Are there any (intelligent) pattern-matching routines in PARI/GP, or
> some other package developed on top of PARI ? Fo example, suppose one 
> want to look for the first occurrence of the pattern:
> 
> 	xyxyxy
> 
> in the first 1-million digits of Pi ... 

Not that I know of.

> I am doing this by brute force right now, but it occurs to me that 
> using something like Boyer-Moore or Knuth-Morris-Pratt may be more
> eficient ...

If you want to work with 10^6 digits, then you can afford to write
the output to a file and use a dedicated program to analyse the output.
Also there are probably more efficient programs than PARI for computing
1-million digits of Pi (and probably you can grab the digits on the net).

Cheers,
Bill.