NAME File::RandomLine - Retrieve random lines from a file VERSION This documentation refers to version 0.19 SYNOPSIS # Fast but biased randomness use File::RandomLine; my $rl = File::RandomLine->new('/var/log/messages'); print $rl->next; print join("\n",$rl->next(3)); # Slow but uniform randomness $rl = File::RandomLine->new('/var/log/messages', {algorithm=>"uniform"}); DESCRIPTION This module provides a very fast random-access algorithm to retrieve random lines from a file. Lines are not retrieved with uniform probability, but instead are weighted by the number of characters in the previous line, due to the nature of the algorithm. Lines are most random when all lines are about the same length. For log file sampling or quote/fortune generation, this should be "random enough". Note -- when getting multiple lines, this module resamples with replacement, so duplicate lines are possible. Users will need to check for duplication on their own if this is not desired. The algorithm is as follows: * Seek to a random location in the file * Read and discard the line fragment found * Read and return the next line, or the first line if we've reached the end of the file * Repeat until the requested number of random lines have been found This module provides some similar behavior to File::Random, but the random access algorithm is much faster on large files. (E.g., it runs nearly instantaneously even on 100+ MB log files.) This module also provides an optional, slower algorithm that returns random lines with uniform probability. USAGE "new" $rl = File::RandomLine->new( "filename" ); $rl = File::RandomLine->new( "filename", { algorithm => "uniform" } ); Returns a new File::RandomLine object for the given filename. The filename must refer to a readable file. A hash reference may be provided as an optional second argument to specify an algorithm to use. Currently supported algorithms are "fast" (the default) and "uniform". Under "uniform", the module indexes the entire file before selecting random lines with true uniform probability for each line. This can be significantly slower on large files. "next" $line = $rl->next(); @lines = $rl->next(5); ($line1, $line2, $line3) = $rl->next(); Returns one or more lines from the file. Without parameters, returns a single line if called in scalar context. With a positive integer parameter, returns a list with the specified number of lines. "next" also has some magic if called in list context with a finite length list of l-values and will return the proper number of lines. INSTALLATION The following commands will build, test, and install this module: perl Build.PL perl Build perl Build test perl Build install BUGS Please report bugs using the CPAN Request Tracker at http://rt.cpan.org/NoAuth/Bugs.html?Dist=File-RandomLine AUTHOR David A Golden http://dagolden.com/ COPYRIGHT Concept and code for "magic" behavior in array context taken from File::Random by Janek Schleicher. All other code Copyright (c) 2005 by David A. Golden. This program is free software; you can redistribute it and/or modify it under the same terms as Perl itself. The full text of the license can be found in the LICENSE file included with this module. SEE ALSO * File::Random * "Re^2: selecting N random lines from a file in one pass", perlmonks.org, static URL: http://perlmonks.thepen.com/417065.html