http://qs321.pair.com?node_id=11119066


in reply to Perl::Critic says don't modify $_ in list functions and other things

Geez. At the first block of code, why not:
my @list = map { s/\r\n$//; $_} <$fh>; my @uc_list = map { s/\r\n$//; uc $_; $_} <$lc_fh>; my @split_list = map { s/\r\n$//; split(/\|/, $_); [@$_] } <$piped_fh> +;
UPDATE: I have found some rare issues using chomp() in multi-platform situations. chomp() does a great job when dealing with files generated upon that particular platform. I have written code that is tested on my Windows box and on a Unix box and then a user uses an old Mac to edit the file. chomp() usually works, but if old Mac formatting is allowed, that is not sufficient. To be honest, the last time I got an error report involving this, I told the old Mac user to set his editor to save files as DOS format. I did not change my code to allow old Mac.

The easiest way to make sure that you delete all of the potential \r \n characters is to remove all white space at the end of the line. I like the suggestion from ikegami of s/\s+\z//. That does have a potential issue with a tab separated line containing a blank field at the end of the line.