utf8 bloody chars

[thorvald]

Also new in perl. I'm following "Learning Perl" from O'Reilly and want to print greek letters. Here's what I do:

use 5.010;
use utf8;
$pi_char = chr(OxO3c9);
print "Greek letter here $pi_char\n";

However, in the output I get:

Greek letter here

[bitingduck]

Be careful about zeros vs. the letter "O". If I put zeros where you have "O"s I get a lower case omega. It's sometimes hard to tell when reading, but the letter tends to be a bit rounder and the number more ovalized. Back in the day publishers were better about using zeros with slashes through them- I don't recall if O'Reilly has moved away from that.

[thorvald]

Thanks so much to both of you. I had never use utf before, since I have only done numerics so far

[AnomalousMonk]

Especially if you are a beginner, you should place the statements  use warnings; and  use strict; at the beginning of all your code:

c:\@Work\Perl\monks>perl -le
"use warnings;
 use strict;
 ;;
 use utf8;
 ;;
 my $pi_char = chr(OxO3c9);
 print qq{Greek letter '$pi_char' here};
"
Bareword "OxO3c9" not allowed while "strict subs" in use at -e line 1.
Execution of -e aborted due to compilation errors.

I.e., "OxO3c9" (letter "O" vice decimal digit "0") is not a hex number, as already noted. See warnings and strict.

In other words, "Help Perl to help you."

[shmem]

To interpolate unicode chars into strings you can use also

print "Greek letter here \x{03c9}\n";
print "Greek letter here \N{U+03c9}\n";
print "Greek letter here \N{GREEK SMALL LETTER OMEGA}\n";

as stated in perlunicode. If you get the warning Wide character in print, use the -C command line switch at least as -CO (that's an 'oh', not a zero :-). See perlrun.

You can find official unicode character names at decodeunicode.

[AppleFritter]

You can find official unicode character names at decodeunicode.

I've also found McLean's Unicode Character Finder very helpful when looking for specific characters.

[moritz]

use the -C command line switch

Or binmode STDOUT, ':encoding(UTF-8)'; inside the script.

[Tux]

Another remark: the use utf8; is only required if the source code contains literal utf-8:

$ perl -CO -wE'say "\x{03c9}"'
ω
$ perl -CO -wE'say "ω"'
Ï
$ perl -CO -Mutf8 -wE'say "ω"'
ω