A Guide for the Survival of Humankind, and Improving the World, Society, and Yourself!

Popular Articles

Reasons For and Against
Human Cloning

Globalization of the USA

The Pros and Cons of
Capital Punishment

World's Stupidest Game

Latest Articles

Corporal Punishment

Garage Door Confusion

Childhood is Stressful

Chinese Saying about Children

Subscribe to

Main Menu

HOME

SAVE HUMANITY (50)

HELP THE WORLD (100)

IMPROVE GOVERNMENTS (90)

PERSONAL DEVELOPMENT (248)

COMPUTER HELP (126)

ARCHIVE (all articles)

SEARCH THIS SITE

CONTACT US

DISCLAIMER

PLEASE HELP: DONATE

UNIX Support: Recursively GREP

I've always had two major problems with the UNIX "grep" command. First, the "grep" command will gladly read a non-text file, and thus produce garbage if the grep pattern was found. Second, the "grep" command can be used to search in all the files in a single directory, but the command will not recursively search all sub-directories. As a result, I wrote a Perl script, called recursive grep or "regrep.pl", to resolve these issues.

Here is the syntax to run this script.

*$ regrep.pl* <WORD> [PATH1] [PATH2] ... [OPTIONS]**

The first argument is the only required argument for the value to search or grep for. The optional "PATH" arguments that can be next are the directories where to start to recursively search for text files to grep. The default path is the current directory. Finally, the last optional argument is for these options:

"-verbose" displays the matched lines for each file searched. "-list" lists the files that have matching text. "-edit" opens the matching files in an editor. "-nocase" performs a case insensitive search.

Alternatively, you can just use the "-help" argument to see the usage and some examples of how to use this script.

You can download this Perl script here, or you can copy-n-paste from down below. I would like to thank Monty Scroggins for enhancing this script with more options and teaching me a bit more Perl.

#!/usr/bin/env perl
 
 ###############################################################################################
 # This script was written for www.philforhumanity.com.
 # If you have any questions, issues, or enhancement requests, please contact us via the website.
 # Use this script at your own risk.
 ###############################################################################################
 
 $|=1;    # set output buffering to off
 $[=0;    # set array base to 0
 $,=" ";  # set output field separator
 $\="\n";  # set output record separator
 
 use Getopt::Long;
 use File::Basename;
 
 my ($list,$edit,$verbose,$nocase,$help,$debug);
 my $editor = "kate";
 
 # Get arguments
 GetOptions (
 'list'       => \$list,
 'edit'       => \$edit,
 'verbose'    => \$verbose,
 'nocase'     => \$nocase,
 'help'       => \$help,
 'debug'      => \$debug
 );
 
 # If "-help" is the only argument, then print the usage and exit.
 if ($help == 1) 
 {
    &usage;
    exit;
 };
 
 # Check the number of arguments, and exit with usage if no arguments
 my $number_of_arguments = scalar(@ARGV);
 &usage("ERROR: Incorrect number of arguments.") if (scalar(@ARGV)<1);
 
 # If "-nocase" argument is used, then see flag to "-i" for grep, otherwise set flag to nothing.
 my $nocaseflag = (defined $nocase ? '-i' : '');
 
 # Define query word
 my $grep_query = $ARGV[0];
 
 # If more than one argument, then collect the paths to search in
 my @paths_to_search = (".");  # Default is the current directory
 if ($number_of_arguments > 1)
 {
   # Remove first argument, since it is not a path and we collected that data already.
   shift @ARGV;
 
   # Save paths to a meaningful variable name, and overwrite default value
   @paths_to_search = @ARGV;
 }
 
 # Query each path for files only
 my @list_of_files = ();
 my $path;
 foreach $path (@paths_to_search)
 {
   @list_of_files = (@list_of_files, `find $path -type f -name '*' -print`);
 }
 
 # For each file found, find which files are text files and then grep only text files
 my $file;
 my $find_output;
 my $found = 0;
 my @grep_output;
 my $line;
 my @foundfiles;
 my $matchcount=0;
 chomp @list_of_files;
 foreach $file (@list_of_files)
 {
   #skip editor backup files, or zip files
   next if ($file =~ /(~|\.bak|\.zip|\.gz)$/);
 
   # Determine the type of each file using the file command in UNIX
   @grep_output = `file $file`;
   $find_output = @grep_output[0];
 
   # If the file type is text, then grep this file, else do not get this file
   #   NOTE: Oddly, zip files are text files too.
   if ($find_output =~ /text/)
   {
     # Grep for data (-n is for line numbers)
     @grep_output = `/bin/fgrep -n $nocaseflag $grep_query "$file"`;
 
     # Output data if grep found something
     if (@grep_output)
     {
      chomp(@grep_output);
 	   if ($verbose) 
 	   {
 		   print "==================================================================";
 		   print "File: $file";
 	   }
 	   push(@foundfiles,$file);
      $matchcount += scalar(@grep_output);
      foreach $line (@grep_output)
      {
 	     print $line if ($verbose);
      }
      $found = 1;
     }
   }
 }
 
 # Print ending delimiter if anything was found
 if ($found == 1)
 {
   my $matchfiles = scalar(@foundfiles);
   print "==================================================================" if ($verbose);
   print qq{$matchcount matches in $matchfiles files};
 
   #if the --list option was specified, display a list of all the matching files
   if ($list) 
   {
 	  print qq{Matched files:};
 	  foreach (@foundfiles) 
 	  {
 		  print "   $_";
 	  }
   }
   if ($edit)
   {
 	  `$editor @foundfiles`;
   }
 }
 
 sub usage 
 {
    my $message = shift;
    print qq{\n$message} if ($message);
    my $script = basename($0);
    print qq{
    USAGE: $script <word> [path1] [path2] ... [options]
 
    options:
     -verbose - display the matched lines for each file searched.
     -list    - list the files that have matching text.
     -edit    - open the matching files in an editor.
     -nocase  - performs a case insensitive search.
 
    The default path is the current directory.
    Spaces must be escaped to be included in the search string.
    Files ending in '~', '.zip' or '.gz' are automatically ignored.
 
    examples:
     $script 'Time\\ Clock' --list (searches files for the string "Time Clock" and displays a list of files that have matching text)
     $script Groucho --edit (searches files for the string "Groucho" and opens the files in an editor)
     $script Groucho --edit --nocase (searches files for the string "Groucho" (case insensitive) and opens the files in an editor)
     $script Harpo   --verbose (searches files for the string "Harpo" and displays all the matches to the screen)
    };
    exit 1;
 }#sub
 
 # Exit
 exit 0;

by Phil B.

Related Articles » Generate Random Numbers » ClearCase Support: How to Find Checked Out Elements » ClearCase Support: Common ClearCase Labeling Error Message

A Guide for the Survival of Humankind, and Improving the World, Society, and Yourself!

UNIX Support: Recursively GREP

Related Articles » Generate Random Numbers » ClearCase Support: How to Find Checked Out Elements » ClearCase Support: Common ClearCase Labeling Error Message

Related Articles
» Generate Random Numbers
» ClearCase Support: How to Find Checked Out Elements
» ClearCase Support: Common ClearCase Labeling Error Message