Alan writes..
>"Kenneth Jideofor [ MTN - Ikoyi ]" wrote:
>>
>> Hi folks,
>>
>> I have two files, FILE1 and FILE2. Each of the two files is an array
>> of numbers. I want to check for the occurrence of each element of
>> FILE1 in FILE2. In other words, FILE2 is checked for the occurrence
>> of each of the elements in of FILE1. For any element of FILE1 that
>> occurs in FILE2, this element is saved in a file, FILE3.txt.
>>
>> I tried the attached Perl script to perform this exercise but the
>> script failed with the following error message:
>>
>> "main:database" used only once: possible typo at ./GREP1.pl line 8"
>>
>> Kindly assist me with a solution to actualising this task.
>
>try this:
>
>#!/usr/bin/perl -w
>
>open FILE1, "</tmp/AAA.txt" or die "Can,t open AAA.txt: $!";
>open FILE2, "</tmp/DATA.dat" or die "Can't open DATA.dat: $!";
>open OUT, ">>/tmp/FILE3.txt" or die "Can't open BBB.txt: $!";
>
>my @numbers = <FILE1>;
>my @database = <FILE2>;
>close(FILE1);
>close(FILE2);
>
>foreach $number (@numbers) {
> chomp($number);
> print OUT grep {/$number/} @database;
>}
>close(OUT);
What if FILE1 contains the number 1 and FILE2 does not contain 1 but
contains 11 then the regex will match against 11 and print '1' out into
FILE3 but the number 1 was not in FILE2 at all.
Additionally your syntax:
print OUT grep {/$number/} @database;
Will mean that your output will print out multiple entries into FILE3
for each iteration, eg if FILE1 contained:
1
2
3
And FILE2 contained:
11
12
13
14
21
44
22
Then your output would be as follows (I've added the spaces in for
readability - in fact there would be no spaces or newlines in your
output):
11 12 13 14 21 12 21 22 13
^^^^^^^^^^^
First pass
^^^^^^^^^^^
Second pass
^^
Third pass
At the very least you'd have to do {/^$number$/} but that's MUCH less
efficient than doing { $_ == $number } - perl strips off the \n itself
when converting $_ to a number. But my preference is to use the hash
lookup detailed in my other post.
--
Jason King
_______________________________________________
ActivePerl mailing list
[EMAIL PROTECTED]
To unsubscribe: http://listserv.ActiveState.com/mailman/mysubs
