Perl Guru Forums: Perl Programming Help: Beginner: Using of sed

CGI/Perl Guide | Learning Center | Forums | Advertise | Login

Site Search: in


	MAIN INDEX	SEARCH POSTS	WHO'S ONLINE	LOG IN

Home: Perl Programming Help: Beginner:
Using of sed

ticketE
Novice

Mar 20, 2013, 10:54 AM

Post #1 of 4 (166 views)

Using of sed	Can't Post

Hi,
I want to replace in my text a string which synonim is already present in my hash %synonimous, but my code doesn't function.
What is it wrong?

Code

foreach my $key (keys %synonimous) { 
	`sed -e 's/$key/$synonimous{$key}/' $filename1` > $filename1; 
}

BillKSmith
Veteran

Mar 20, 2013, 12:23 PM

Post #2 of 4 (160 views)

Re: [ticketE] Using of sed [In reply to]	Can't Post

Your code does not work because the second back-tic is in the wrong place.

I would not recommend fixing it. Back-tics are used when you need the standard output. In this case, it is already redirected to a file. The function system would be a better choice.

Better yet, implement your editing in perl.

Code

  
#!perl -p 
use strict; 
use warnings; 
BEGIN{ 
    use vars qw(%SYNONYMS $SOURCE); 
    %SYNONYMS = ( 
        old1 => 'new1', 
        old2 => 'new2', 
        old2 => 'new3', 
    ); 
    my $source  = join '|', keys %SYNONYMS; 
    $SOURCE  = qr/$source/; 
} 
s/($SOURCE)/$SYNONYMS{$1}/e;

USAGE:

Code

perl script.pl infile > outfile

Good Luck,
Bill

ticketE
Novice

Mar 21, 2013, 7:56 AM

Post #3 of 4 (150 views)

Re: [BillKSmith] Using of sed [In reply to]	Can't Post

My problem is that I have to do all in one script and execute this with a command:
./script.perl

It's not correct for my application, I can't change the content of the hash, I only need to find and replace, using my hash and a file of input/output.

BillKSmith
Veteran

Mar 21, 2013, 9:31 AM

Post #4 of 4 (147 views)

Re: [ticketE] Using of sed [In reply to]	Can't Post

I do not think that you understood my code. I have made minor changes to try and answer your objections. Readonly is not required. I use it to prove that neither the hash nor Regex is ever modified.

You seem to be using UNIX. I changed the #! line to the UNIX default. The -p runtime option (on that line) handles all I/O and looping much like sed.

The perl substitution operator is much more powerful than sed's. Looping through all the keys is done by the Regex itself. The match is stored in the special variable $MATCH (Refer: perldoc perlvar). Note the /e on the operator. It tells perl to eval the resulting hash expression before making the final substitution.

Code

#!\usr\local\bin\perl -p 
# Note the runtime option -p  (Refer: perldoc perlrun) 
use 5.010;                              # Needed for \\= 
use strict; 
use warnings; 
use Readonly; 
use English; 
BEGIN{ 
    $ARGV[0] //= 'input_file_name';     # Set default file name 
    use vars qw( %SYNONYMS $SOURCE );   # Needed outside scope of BEGIN 
    Readonly::Hash %SYNONYMS => (       # Make sure hash is never changed 
        old1 => 'new1', 
        old2 => 'new2', 
        old3 => 'new3', 
    ); 
    my $source  = join '|', keys %SYNONYMS; 
    Readonly::Scalar $SOURCE  => qr/\b$source\b/o;  # Compile Regex only once 
} 
s/$SOURCE/$SYNONYMS{$MATCH}/eg;     # Edit every line (Each line in $_)

Good Luck,
Bill

Search for (options)

Web Applications & Managed Hosting Powered by Gossamer Threads
Visit our Mailing List Archives