Inserting new column

Avi_Perl_Beginner has asked for the wisdom of the Perl Monks concerning the following question:

My requirement is to read all.csv files in the format groupid_groupname.csv in a folder. Secondly after reading create two new columns with groupid & groupname and print the constant value fetched above. I have written the code but while using just print it works perfect and shows exact result but when writing to csv file..it goes for a toss. Please advice

use strict;
use warnings;
use Text::CSV;
use Text::CSV_XS;
use autodie;

print_Filenames_directory();


sub print_Filenames_directory
{
my @files = <*.csv>;
    foreach my $file (@files)
    {
        open my $in, "+<:encoding(utf8)", $file or die $!;
        
        my $csv = Text::CSV_XS->new ({ binary => 1, auto_diag => 1 });
                                
        my @name = split _,$file;
        my $grpid = $name[0];
        my $grpname = substr($name[1], 0,-4);
        
        
        my @headers; 
    
        while (my $row = $csv->getline($in)) 
        {      
             if($.==1) 
            {
               # Add value to @columns array
               splice @$row, 0, 0, "Group ID";
               splice @$row, 1, 0, "Group Name";                
               @headers = "@$row\n";
               print  "@headers";
               #$csv->print(@headers); 
            } 
            else
            {
               splice @$row, 0, 0, $grpid;
               splice @$row, 1, 1, $grpname;                
               print "@$row\n"; 
               #$csv->print($in, @$row);                              
+              
            }   
                               
        }
        # check for errors
        $csv->eof or $csv->error_diag();
            
        close $in;
    }

}
[download]

Comment on Inserting new column Download Code

Replies are listed 'Best First'.
Re: Inserting new column by hippo (Archbishop) on May 27, 2015 at 10:42 UTC
You cannot (should not) write to the open file while you are still reading from it. Open a new tempfile, write to that, then once the whole file has been read and written, close both and rename the tempfile as the original file (taking a backup first to be prudent). The alternative is to read the whole file in first, store the results in RAM and then write them all out in one go. HTH.	[reply]
Re: Inserting new column by Tux (Canon) on May 27, 2015 at 11:04 UTC
Too many thing wrong to comment on all. Here is a simplified version which I think does what you want: use strict; use warnings; use autodie; use Text::CSV_XS 1.18; print_Filenames_directory (); sub print_Filenames_directory { my $csv = Text::CSV_XS->new ({ binary => 1, auto_diag => 1 }); foreach my $file (glob "*.csv") { open my $fh, "<", $file or die $!; my @name = split m/_/ => $file; my $grpid = $name[0]; my $grpname = substr $name[1], 0, -4; my @headers = ("Group ID", "Group Name", @{$csv->getline ($fh) +}); my @rows; while (my $row = $csv->getline ($fh)) { # update: forgot to remove the two unwanted splice lines push @rows, [ $grpid, $grpname, @$row ]; } close $fh; open $fh, ">", $file or die $!; $csv->say ($fh, \@headers); $csv->say ($fh, $_) for @rows; close $fh; } } [download] Enjoy, Have FUN! H.Merijn	[reply] [d/l]
Re^2: Inserting new column by Avi_Perl_Beginner (Initiate) on May 27, 2015 at 12:29 UTC
TUX : Thanks a lot...it was excellent one time solution	[reply]
Re: Inserting new column by Anonymous Monk on May 27, 2015 at 10:41 UTC
`open my $in, "+<:encoding(utf8)"` You know utf8 is not the same as UTF-8 `use autodie; ... or die $!;` hah :) `my @name = split _,$file; my $grpid = $name[0] ; my $grpname = substr($name[1], 0,-4);` see see perlintro#Parentheses for capturing and http://perldoc.perl.org/perlrequick.html#Extracting-matches while...if($.==1) ... No need to put the first getline into the loop, just do it before the loop `#$csv->print($in, @$row);` You need an $outcsv because Modifying an existing file is three steps :) read old file create a new file rename new file to name of old file	[reply] [d/l] [select]