Stanislav Nedelchev wrote:
Rob Dixon wrote:
Andrej Kastrin wrote:
Dear all,
Question about the sum function; the file structure is as follows:
A|100
A|200
A|150
B|20
B|90
C|10
C|30
C|300
The result I want to obtain is to sum values in the second column
(columnB) for each particular letter in the first column (ColumnA);
e.g.:
A|450
B|100
C|330
I don't want to use hash structure because the input file is very
large. Is there any simple way to do that step-by-step: to sum up
values in columnB until the letter in columnA changes and print the
result...
Thanks in advance for any suggestion, Andrej
#!/usr/bin/perl
use strict;
use warnings;
open FH1, "< test.txt" or die "Can't open file : $!";
while (<FH1>) {
chomp;
($columnA,$columnB)=split /\|/;
... ???
}
my ($label, $total);
while (<FH1>) {
chomp;
my ($columnA,$columnB)=split /\|/;
if ($label and $columnA ne $label) {
print "$label|$total\n";
$total = 0;
}
$label = $columnA;
$total += $columnB;
print "$label|$total\n" if eof;
}
HTH,
Rob
It's was interesting to see some different solutions of the problem .
I was very usefull to me but i think that if data is in random order it
will not work in expected way.
Maybe I'm wrong.
Data :
A|100
B|20
A|200
A|150
B|20
B|90
C|10
C|30
C|300
Result:
A|100
B|20
A|350
B|110
C|340
Best Regards.
I wrote this solution specifically because no one else had answered the original
poster's question:
Andrej Kastrin wrote:
I don't want to use hash structure because the input file is very
large. Is there any simple way to do that step-by-step: to sum up
values in columnB until the letter in columnA changes and print the
result...
This program does exactly that, although I would have coded a hash solution
if I had the choice.
Rob
--
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
http://learn.perl.org/
