Hello all:

So as not to further distract the cb from its discussion of Zappa's life's work, I bring this question here.

I am working on a script to clean and import a large csv data file into a postgres backend. My duplicate checks are failing to prevent fatal errors related to unique data constraints applied to some tables. As I drill into this issue, what I'm finding is that my regex's are not working as I thought they would.

#!/usr/bin/perl -w
use strict;
use warnings;

open ('CSV','<',$file);
while (<CSV>){
  $counter++;
  @fields = split ",",$_;
  my($ndx);
  foreach $ndx (0 .. (0 + @fields -1)){
    $fields[$ndx] =~ s/\s*$//g;
    $fields[$ndx] =~ s/'/\\'/g;
    $fields[$ndx] =~ s/"//g;
    if($ndx == 32){
      print STDERR "For field $ndx, the value is now: |$fields[$ndx]| 
+\n";
      my @chars = split(//, $fields[$ndx]);
      foreach my $chr (@chars) {
        print STDERR '|' . ord($chr) . '|,';
      }
      print STDERR "\n";
    }
  }

close('CSV');
[download]

This is yielding debug output which looks like this:

For field 32, the value is now: |096  |
|48|,|57|,|54|,|32|,|32|,
[download]

Field #32 is only a representative sample. This issue is showing up on several fields actually.

Can anyone please explain why this is happening for me, please? And more important what I can do about it?

Thanks,

-- Hugh