in reply to Re^3: Processing data in on dir and copying, then processing into another dir
in thread Processing data in on dir and copying, then processing into another dir

Hi Zentara, I am using absolute paths. I have set up to testdir for testing this script. When I run the script the stdout is: GLOB(0x804cbdc) . .. RESULTS 2008 RESULTS 2009 RESULTS 2007 The glob should be the files that match from grep, but I nothing is being moved from 'testdir1' to 'testdir2'. 
#!/usr/bin/perl
use warnings;
use strict;

# goes from current dir, and looks thru
# a subdir named 7, for subdirs starting with
# Jan_2009_,  then goes into that subdir and finds
# files begiining with a or b or c
# making your particular regexes is up to you
#opendir ( my $rs, "tarfs-orion/orion/Data/Sequencing_Results/RESULTS 
+2009");
my $topdir = "/home/mgavi/testdir1";
my @subdirs = get_sub_dirs($topdir);
my $newdir = "/home/mgavi/testdir2";

sub get_sub_dirs {
    my $dir = shift;
    opendir my $dh, $dir or die "Error: $!";
    print "$dh\n";
    my @dirs =  readdir $dh;
    print "@dirs\n";
    @dirs = grep /^Results(.*)$/,@dirs;
    closedir $dh;
    return @dirs;
}

print "@subdirs\n";

my @files;
foreach my $dir(@subdirs){
    opendir my $dh, "$topdir/$dir" or die "Error: $!";
    @files = readdir $dh;
    print "@files\n";
    @files = grep /^(\d+\D\d)_(LacZ|pgK|SD|SU)(.*)$/,@files;
    foreach my $file(@files){
    system( "mv $file $newdir/$file");
    }
    closedir $dh;
}

print "@files\n";

[download]
  • Comment on Re^4: Processing data in on dir and copying, then processing into another dir
  • Download Code

Replies are listed 'Best First'.
Re^5: Processing data in on dir and copying, then processing into another dir
by zentara (Cardinal) on Jan 11, 2009 at 14:30 UTC
    Yeah, you have path issues, and possibly regex issues. mv requires full path names. The following works. In testdir1 I have 2 subdirs Results_2008 and Results_2009. In those subdirs , I have files: 
    Results_2008:
2008_12112L5_LacZ  2008_12112L5_pgK  2008_12112L5_SD  2008_12112L5_SU

Results_2009:
2009_12112L5_LacZ  2009_12112L5_pgK  2009_12112L5_SD  2009_12112L5_SU

    [download]
    this script moves them all, and notice the regex(probably can be improved) and full path in the system mv 
    #!/usr/bin/perl
use warnings;
use strict;

my $topdir = "/home/zentara/testdir1";
my @subdirs = get_sub_dirs($topdir);
my $newdir = "/home/zentara/testdir2";

sub get_sub_dirs {
    my $dir = shift;
    opendir my $dh, $dir or die "Error: $!";
#    print "$dh\n";  #this is your glob
    my @dirs =  readdir $dh;
    @dirs = grep /^Results(.*)$/,@dirs;
    closedir $dh;
    return @dirs;
}

print "subdirs = @subdirs\n";

foreach my $dir(@subdirs){
    opendir my $dh, "$topdir/$dir" or die "Error: $!";
    my @files = readdir $dh;
    print "raw files = @files\n";
    @files = grep /^[\d]+_[\dL]+_(LacZ|pgK|SD|SU)(.*)$/,@files;
    print "$dir -> @files\n";

    foreach my $file(@files){
     system( "mv $topdir/$dir/$file $newdir/$file");
      
    }
    closedir $dh;
}

    [download]
    Also remember, if you want to separate your files out into subdirs, you need to make the subdirs first, before trying to mv them there. See cloning a directory tree
    I'm not really a human, but I play one on earth Remember How Lucky You Are
[reply]
[d/l]
[select]
      My first step is to get the files in Results 2009 that match the regex to mv to a new dir. Next, in the new dir I want to match all the file with dir that have the same id number.
      Ex. 2009_12112L5_LacZ 2009_12112L5_pgK 2009_12112L5_SD 2009_12112L5_SU >
      12112, so each of the files will go into a new dir called 12112/12112_sa.
      I think a hash should help me with the matching. Much Thanks!
[reply]
      Did I mention that OS is Unix? Also after testing the code it returns:

      GLOB(0x804cbdc)
      . .. RESULTS 2008 RESULTS 2009 RESULTS 2007
      subdirs =
      nothing is mv to newdir
[reply]

In Section Seekers of Perl Wisdom