Find last occurrences of a string from multiple files in a directory using perl - perl

Question

In a directory have multiple file . from all that file i have to find the last occurrence of a string and print that line into a result.txt file.
Let assume the files are like demo*.txt.
And the string i have to find is "FREQ" .
The code i have tried is as follows :
for ( glob("demo*.txt" ) ) {
my $xm = $_;
open my $fh, '<', $xm;
my $pat = "FREQ" ;
while(my $asa = <$fh>) {
my #last = grep(/FREQ/, $asa);
}
}
print " grep : #last\n";
I have tried this code but this giving nothing and I cant able to find the mistake i am making.

Answer 1

As #vkk05 suggests, you need to add use strict & use warnings to your code
use strict;
use warnings;
for ( glob("/tmp/demo*.txt" ) ) {
my $xm = $_;
open my $fh, '<', $xm;
my $pat = "FREQ" ;
while(my $asa = <$fh>) {
my #last = grep(/FREQ/, $asa);
}
}
print " grep : #last\n";
When I run that I get
Possible unintended interpolation of #last in string at /tmp/fred.pl line 15.
Global symbol "#last" requires explicit package name (did you forget to declare "my #last"?) at /tmp/fred.pl line 15.
Execution of /tmp/fred.pl aborted due to compilation errors.
The problem is the my #last line. You are defining #last in the scope of the while loop. That means the variable #last gets deleted once the while loop exits.
Moving the definition of #last to outside the while scope and moving the print line to run immediately after the while loop gives this code
use strict;
use warnings;
for ( glob("/tmp/demo*.txt" ) ) {
my $xm = $_;
open my $fh, '<', $xm;
my $pat = "FREQ" ;
my #last;
while(my $asa = <$fh>) {
#last = grep(/FREQ/, $asa);
}
print " grep : #last\n";
}
and assuming a test file /tmp/demo1.txt
$ cat /tmp/demo1.txt
abc
FREQ 123
FREQ 456
running the script gives
grep : FREQ 456
Finally, the use of grep in this context is confusing -- it is intended to be used when iterating through a list of values, rather than a single entry as in this case. Although the code works fine with the way you are using grep, a more perl-idiomatic solution that uses grep would be
use strict;
use warnings;
my $pat = "FREQ" ;
for my $xm ( glob("/tmp/demo*.txt" ) ) {
open my $fh, '<', $xm
or die "Cannot open '$xm': $!";
my #matches = grep { /$pat/ }
<$fh>;
print " file $xm : $matches[-1]\n";
}
Points to note
grep is used here in a proper list context, along with <$fh> to walk the file and match all lines that include the pattern stored in $pat
The results are stored in #matches, so to get the last one, the print statement uses $matches[-1] to index the final entry in #matches.

Answer 2

If this is a one time only task I would use a simple one-liner like this.
Create four test files. The last is empty:
echo -e "FREQ\nline 2\nFREQ again\nline4" > demo1.txt
echo -e "FREQUENT\nline 2 in demo2\nShrek\nXFREQ here" > demo2.txt
echo -e "nothing to see\nnothing to see!" > demo3.txt
touch demo4.txt
Run:
perl -E '#f=#ARGV; /FREQ/ and $L{$ARGV}=$_ while<>; print "Last FREQ line in $_: ".($L{$_}//"<none>\n") for #f' demo*.txt | tee result.txt
Output:
Last FREQ line in demo1.txt: FREQ again
Last FREQ line in demo2.txt: XFREQ here
Last FREQ line in demo3.txt: <none>
Last FREQ line in demo4.txt: <none>