You can try hexdump -C script.pl to inspect it even further.

Chances are that your system's nroff (or whatever manpage processor you're using) substituted some quotes with a "fancy" quote that now causes trouble.

The encoding is 'utf-8'. I think you're right, it's the quotes. See below, the quotes are replaced by "..." in hexdump's ASCII, which are 'e2 80 99'. Now where did I put that nroff book...

00000000  75 73 65 20 4d 49 4d 45  3a 3a 4c 69 74 65 3b 0a  |use MIME:
+:Lite;.|
00000010  23 23 23 20 43 72 65 61  74 65 20 61 20 6e 65 77  |### Creat
+e a new|
00000020  20 73 69 6e 67 6c 65 2d  70 61 72 74 20 6d 65 73  | single-p
+art mes|
00000030  73 61 67 65 2c 20 74 6f  20 73 65 6e 64 20 61 20  |sage, to 
+send a |
00000040  47 49 46 20 66 69 6c 65  3a 0a 24 6d 73 67 20 3d  |GIF file:
+.$msg =|
00000050  20 4d 49 4d 45 3a 3a 4c  69 74 65 2d 3e 6e 65 77  | MIME::Li
+te->new|
00000060  28 0a 46 72 6f 6d 20 20  20 20 20 3d 3e e2 80 99  |(.From   
+  =>...|
00000070  6d 65 40 6d 79 68 6f 73  74 2e 63 6f 6d e2 80 99  |me@myhost
+.com...|
00000080  2c 0a 54 6f 20 20 20 20  20 20 20 3d 3e e2 80 99  |,.To     
+  =>...|
00000090  79 6f 75 40 79 6f 75 72  68 6f 73 74 2e 63 6f 6d  |you@yourh
+ost.com|
000000a0  e2 80 99 2c 0a 43 63 20  20 20 20 20 20 20 3d 3e  |...,.Cc  
+     =>|
000000b0  e2 80 99 73 6f 6d 65 40  6f 74 68 65 72 2e 63 6f  |...some@o
+ther.co|
000000c0  6d 2c 20 73 6f 6d 65 40  6d 6f 72 65 2e 63 6f 6d  |m, some@m
+ore.com|
000000d0  e2 80 99 2c 0a 53 75 62  6a 65 63 74 20 20 3d 3e  |...,.Subj
+ect  =>|
000000e0  e2 80 99 48 65 6c 6c 6f  6f 6f 6f 6f 6f 2c 20 6e  |...Helloo
+oooo, n|
000000f0  75 72 73 65 21 e2 80 99  2c 0a 54 79 70 65 20 20  |urse!...,
+.Type  |
00000100  20 20 20 3d 3e e2 80 99  69 6d 61 67 65 2f 67 69  |   =>...i
+mage/gi|
00000110  66 e2 80 99 2c 0a 45 6e  63 6f 64 69 6e 67 20 3d  |f...,.Enc
+oding =|
00000120  3e e2 80 99 62 61 73 65  36 34 e2 80 99 2c 0a 50  |>...base6
+4...,.P|
00000130  61 74 68 20 20 20 20 20  3d 3e e2 80 99 68 65 6c  |ath     =
+>...hel|
00000140  6c 6f 6e 75 72 73 65 2e  67 69 66 e2 80 99 0a 29  |lonurse.g
+if....)|
00000150  3b 0a 24 6d 73 67 2d 3e  73 65 6e 64 3b 20 23 20  |;.$msg->s
+end; # |
00000160  73 65 6e 64 20 76 69 61  20 64 65 66 61 75 6c 74  |send via 
+default|
00000170  0a 0a                                             |..|
00000172
[download]

non-Perl: read my music blog  be my fan at Reverbnation  bump me at the Sixty-One 

Would a simple use utf8; solve your problem? See perldoc utf8.

Jim

Look at all lower-case "a"'s in the line to find those that look unusual.

caret on top? :)

-- Randal L. Schwartz, Perl hacker

I prefer blueberry :P

It's more like a hat really.

I find this very annoying also. It's because the perldoc renderer is doing silly things like replacing regular '-' dash/hyphen characters with chr(226) (instead of the normal chr(45) character) and replacing single ticks with the "prettier" version of the single quote. I have not been able to figure out how to get this to display properly without switching completely to raw or text mode.

My solution was to bind a key in vim to run this silly little script. So whenever I paste some code from some perldoc I'm viewing, I run this over the code.

#!/usr/bin/perl -n

s/‐/-/g;
s/−/-/g;
s/’/'/g;
print;
[download]

I notice it renders oddly in the perlmonks node, but you get the idea. I literally just copy and pasted the offending symbol into this script.

Yes, thanks, you put me right onto "the" answer for my work enviroment.

In your vimrc, put

vmap ,qq :%s/’/'/g<CR>
nmap ,qq :%s/’/'/g<CR>
[download]

where the first quote is entered as a digraph: Ctrl-V Ctrl-K '9.

At that point "comma q q" is mapped in vim to replace all the bad fancy quotes with good single quotes.

non-Perl: read my music blog  be my fan at Reverbnation  bump me at the Sixty-One 

That will only handle that one character. The problem is there are other characters that are modified as well. After my previous post the other day, I ended up going and making a much more verbose version of that script. Now the bad characters can be referred to by name. Plus I added a silly way of making it display the identity of characters to make it easier to find more that need to be fixed.

#!/usr/bin/perl -n

#use strict;
#use warnings;
use charnames ();
use encoding "utf8";

$|++;

my $chars = {
    'HYPHEN'                      => '-',    #  \x{2010}
    'MINUS SIGN'                  => '-',    #  \x{2212}
    'FIGURE DASH'                 => '-',    #  \x{2012}
    'RIGHT SINGLE QUOTATION MARK' => "'",    #  \x{2212}
    'BOX DRAWINGS LIGHT VERTICAL' => '|',    #  \x{2502}
};

# If the first character is an equal sign, skip it and
# display the identity of each remaining characters.
#
if (/^=/) {
    for my $index ( 1 .. length($_) - 1 ) {

        my $char = substr( $_, $index++, 1 );

        print $char . " "
                    . sprintf( "\\x{%04X}", ord($char) )
                    . "\" = '"
                    . charnames::viacode( ord($char) ) 
                    . "'\n" ;

    }

} else {

    for my $cname ( keys %$chars ) {

        my $char = chr( charnames::vianame($cname) );
        s/$char/$chars->{$cname}/g;

    }

    print;

}

[download]

--
naChoZ

Therapy is expensive. Popping bubble wrap is cheap. You choose.