nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

find from perl

herald10

Hi,
I am trying to print the results from a find command. It just displays the file name and not the line in which the pattern was found.

my $findCmd = qq{find $vpath -name "*" -print | xargs grep \Q$pattern\E 2>/dev/null};
my @results = `$findCmd`;
print $findCmd;

Can anyone please help?

Thanks in advance...

Find more posts tagged with

Comments

rajnag

your are trying to print the command insted of the results in your perl script.

Adam Stoller

Try something like this....

use strict;
my $dir = "...to-be-filled-in-by-you...";
my %hash = ();
my $pattern = "...to-be-filled-in-by-you...";
my $errbuf = '';

process_dir($dir, \%hash, $pattern, $errbuf);
foreach my $f (sort(keys(%hash))){
    print "$f:$_\n" foreach(@{$hash{$f}});
}

sub process_dir{
    my($dir, $hashref, $pattern, $errbuf) = @_;
    if (!opendir(DIR, $dir)){
        $errbuf .= "[opendir] '$dir' ($!)\n";
        return;
    }
    my @entries = readdir(DIR);
    closedir(DIR);
    my @subdirs = ();
    foreach (@entries){
        next if /^\.+$/;
        if (-d "$dir/$_"){
            push(@subdirs, "$dir/$_");
        }
        elsif (! -B "$dir/$_"){
            process_file("$dir/$_", $hashref, $pattern, $errbuf);
        }
    }

    process_dir($_, $hashref, $pattern, $errbuf) foreach(@subdirs);
}

sub process_file{
    my($file, $hashref, $pattern, $errbuf) = @_;
    if (!open(IN, "<$file")){
        $errbuf .= " '$file' ($!)\n";
        return;
    }
    while(<IN>){
        chomp;
        next if ($_ !~ /\Q$pattern\E/);
        push(@{$hashref->{$file}}, $_);
    }
    close(IN);
}

--fish
Senior Consultant, Quotient Inc.
http://www.quotient-inc.com

Valentine

wow. I can't believe nobody mentioned find2perl.
There are already perl modules that come with TS.

Instead of running

[ find $vpath -name "*" -print | xargs grep \Q$pattern\E 2>/dev/null ]

try
iw-home/iw-perl/bin/find2perl $vpath -name "*" -print | xargs grep \Q$pattern\E 2>/dev/null

which will generate the perl code you can use in your scripts.

Adam Stoller

Question - did you try running find2perl like that *before* you posted?
I think not.

find2perl is very handy for doing *just* the 'find' part of that command - however, you still need to account for the entire right-hand-side of the pipe by yourself....

--fish
Senior Consultant, Quotient Inc.
http://www.quotient-inc.com

Log3.PNG

Log4.PNG

Valentine

hmm ... not this one. Smiley Wink

But checking the command more closely, I would use
"grep $pattern *" instead of find.

Adam Stoller

Frankly, I'd have used (as my model) find $vpath -type f -print | ... instead of using -name "*"

The reason for using find ... | xargs grep ... over grep ... * is because there is a limit to the number of command line arguments and/or the length of a complete command line that can be processed - xargs does a very nice job of pacelling out the data from find into "legal" command lines that can be passed to grep

Thus, the perl code I wrote, which recurses through all the directories starting with the one passed in, and then attempts to process all [non-binary] files in those directories to see if it can find matches for the pattern being sought. I didn't get a chance to do any benchmarking on it, it might be more efficient to slurp the entire file in at once (either into a single string or an array) and use Perl's built-in grep function - but I'll leave that exercise to someone else.

--fish
Senior Consultant, Quotient Inc.
http://www.quotient-inc.com

gzevin

here us a sample of my code that I wrote some time ago...

@ARGV = qw(.) unless @ARGV;
use File::Find ();
sub find(&@) { &File::Find::find }
*name = *File::Find::name;
find { print $name if -d && $File::Find::name !~ /EDITION/ } @ARGV unless $File::Find::name =~ /EDITION/ ;

Greg Zevin, Ph.D. Comp. Sc.
Independent Interwoven Consultant/Architect
Sydney, AU

laj1

> I am trying to print the results from a find command. It just displays the file name and not the line in which the pattern was found.
> my $findCmd = qq{find $vpath -name "*" -print | xargs grep \Q$pattern\E 2>/dev/null};

OK, despite every other thing that everybody has said so far, the -n switch to grep is what makes grep
print the line on which the match occurred.

my $findCmd = qq{find $vpath -name "*" -print | xargs grep -n \Q$pattern\E 2>/dev/null};

Try it from the command line and see what I mean.

Also, man grep (the manual for the grep command, which lists all of the command line switches for grep) would have been a great place to start.

Having said all that, other folks have suggested using find2perl.exe and/or Find.pm. IW has also provided a directory tree walking module (whose name escapes me at the moment). It may be faster in the long run to keep all your processing, including the regex matching in perl, to avoid all of the overhead of forking and exec'ing grep.

Len.
--
Len Jaffe LenJaffe@JaffeSystems.com
Leonard Jaffe Computer Systems Consulting Ltd.
Columbus, OH, USA

Len Jaffe
My Heart Is A Flower
Update your DevNet profile - let us know who you are!

Adam Stoller

OK, despite every other thing that everybody has said so far, the -n switch to grep is what makes grep print the line on which the match occurred.

Correction - the -n flag is what makes grep print the line number on which the match occurred, grep, without the -n should still print the line on which the match occurred.

--fish
Senior Consultant, Quotient Inc.
http://www.quotient-inc.com