I would use the second approach. I would think that it would be better
performance-wise, not to mention I always like to have a backup in case
things don't work the way I expected.
Alother thing you have to try to do is only read in as much data at one time
as you need to, because a million rows in memory could really kill you.
Maybe try something like this (no doubt someone will have a much better
solution, but I figured I would try anyway):
#!/usr/bin/perl
use strict;
use English;
my $infile = "somefile";
my $outfile = "someotherfile";
open IN, $infile or die $!;
open OUT, "> $outfile" or die $!;
my @buffer = ();
while ( my $line = <IN> ) {
# push the current line on the buffer
# $NR is our current line number in the file
if ( $NR < 4 ) {
# push the row onto the buffer WITH the newline char
push @buffer, $line;
}
# The "4" is our fourth line
elsif ( $NR == 4 ) {
chomp($line);
for ( my $i = 0; $i < @buffer; $i++ ) {
my $bufLine = $buffer[$i];
chomp($bufLine);
# get the adjusted line num for the buffer line
my $lineno = $NR - @buffer + $i;
if ( $lineno == 1 or $lineno == 3 ) {
$bufLine .= $line;
}
# add the line to the output file
print OUT "$bufLine\n";
}
# clear the buffer
@buffer = ();
}
else {
print OUT $line;
}
}
close IN;
close OUT;
Rob
-----Original Message-----
From: Ho, Tony [mailto:[EMAIL PROTECTED]]
Subject: appending to rows
Hi guys
I was wondering if you could help me
I have 2 files.
One file has 3 data rows, A, B and C
The file has 1 data row D.
I need to append row D to rows A and C in the first file, based on some
condition.
Any ideas how I could go about this ?
Another alternative would be for me to open up a new file and write
row 1 : AD
row 2 : B
row 3 : CD
Is the second approach better in terms of performance than the first
approach as I will have more than 1 million rows to deal with ?
Thanks in advance
Tony
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
