You use :raw:encoding($enc):crlf:utf8 for writing.
You use :encoding($enc) for reading.
This lack in symmetry in your IO layers accounts for your lack of symmetry in CRLF handling.
Use :raw:encoding($enc):crlf:utf8 for reading too.
use Data::Dumper qw( Dumper ); sub hexdump { (my $dump = uc unpack 'H*', $_[0]) =~ s/(..)/$1 /g; return $dump; } sub txtdump { local $Data::Dumper::Useqq = 1; local $Data::Dumper::Terse = 1; local $Data::Dumper::Indent = 0; return Dumper($_[0]); } { open(my $fh, '>:raw:encoding(UTF-16le):crlf:utf8', 'test') or die; my $data = "foo\nbar\n"; print("Orig: ", txtdump($data), "\n"); print $fh $data; } { open(my $fh, '<:raw', 'test') or die; local $/; my $data = <$fh>; print("File: ", hexdump($data), "\n"); } { open(my $fh, '<:raw:encoding(UTF-16le):crlf:utf8', 'test') or die; local $/; my $data = <$fh>; print("Read: ", txtdump($data), "\n"); }
Orig: "foo\nbar\n" File: 66 00 6F 00 6F 00 0D 00 0A 00 62 00 61 00 72 00 0D 00 0A 00 Read: "foo\nbar\n"
Update: Added code.
In reply to Re: Module to read - modify - write text files in any unicode encoding
by ikegami
in thread Module to read - modify - write text files in any unicode encoding
by Rudif
| For: | Use: | ||
| & | & | ||
| < | < | ||
| > | > | ||
| [ | [ | ||
| ] | ] |