slurp with pattern match

[bjlockie]

my @checkOutput = ();
if (open( FILE, $inputFilePath . '/checkOutput' )) {
@checkOutput = <FILE>;
close( FILE );
}

I want to slurp only lines that match a pattern.
Can I still slurp or do I need to loop each line?

[Rainer Weikusat]

bjlockie <bjlo...@lockie.ca> writes:
> my @checkOutput = ();

Perl variables don't need to be initialized explicitly. In some pretty
accidental looking circumstances, perl might print warnings when an
automatic conversion of a scalar without a value to a string or an
integer occurs. But even this doesn't apply here because the value of
@checkOutput is an empty array. There's no point in emptying that by
assigning an empty list to it.

You could use grep here:

@checkOutput = grep { /pattern/ } <FILE>;

[J. Gleixner]

Many ways..

Using File::Slurp.

use File::Slurp;
my $file = "$inputFilePath/checkOutput";
my @checkOutput = grep { /blah/ } read_file( $file );


or maybe the file is really large and you don't want to 'slurp'..

my $file = "$inputFilePath/checkOutput";
open( my $out, '<', $file ) || die "Can't open $file: $!";
my @checkOutput;
while( <$out> )
{
push( @checkOutput, $_ ) if /blah/;
}
close( $out );

or make that while a bit more compact..

my @checkOutput = grep { /blah/ } <$out>;

or if you're on a Unix machine and your pattern can be done
using grep from the OS.

open( FH, "/bin/grep 'blah' $inputFilePath/checkOutput |" )
|| die "Can't grep $file: $!";
my @checkOutput = <FH>;
close( FH );

[Ben Morrow]

Quoth "J. Gleixner" <glex_n...@qwest-spam-no.invalid>:

> On 09/20/12 14:56, bjlockie wrote:
> >
> > I want to slurp only lines that match a pattern.
> > Can I still slurp or do I need to loop each line?

<snip>

>
> or maybe the file is really large and you don't want to 'slurp'..
>
> my $file = "$inputFilePath/checkOutput";
> open( my $out, '<', $file ) || die "Can't open $file: $!";
> my @checkOutput;
> while( <$out> )
> {
> push( @checkOutput, $_ ) if /blah/;
> }
> close( $out );
>
> or make that while a bit more compact..
>
> my @checkOutput = grep { /blah/ } <$out>;

This variant reads all the lines of the file into a list before passing
that list to grep, so it isn't suitable for huge files either.

The most memory-efficient option is

while (<$out>) {
/blah/ or next;
# process the line immediately, then throw it away
}

Ben