Re: Copious warnings from Sys::Syslog

[p5sagit/p5-mst-13.2.git] / lib / encoding.pm

diff --git a/lib/encoding.pm b/lib/encoding.pm
index 94ee323..441be33 100644 (file)
--- a/lib/encoding.pm
+++ b/lib/encoding.pm
@@ -1,7 +1,16 @@
 package encoding;
 
+our $VERSION = '1.00';
+
 use Encode;
 
+BEGIN {
+    if (ord("A") == 193) {
+       require Carp;
+       Carp::croak "encoding pragma does not support EBCDIC platforms";
+    }
+}
+
 sub import {
     my ($class, $name) = @_;
     $name = $ENV{PERL_ENCODING} if @_ < 2;
@@ -43,10 +52,15 @@ encoding - pragma to control the conversion of legacy data into Unicode
 
     print "tera\n" if ord(pack("C", 0xdf)) == 0x3af;
 
-    # but pack/unpack are not affected, in case you still
+    # ... as are eq and cmp ...
+
+    print "peta\n" if "\x{3af}" eq  pack("C", 0xdf);
+    print "exa\n"  if "\x{3af}" cmp pack("C", 0xdf) == 0;
+
+    # ... but pack/unpack C are not affected, in case you still
     # want back to your native encoding
 
-    print "peta\n" if unpack("C", (pack("C", 0xdf))) == 0xdf;
+    print "zetta\n" if unpack("C", (pack("C", 0xdf))) == 0xdf;
 
 =head1 DESCRIPTION
 
@@ -75,16 +89,30 @@ since the C<\xDF> on the left will B<not> be upgraded to C<\x{3af}>
 because of the C<\x{100}> on the left.  You should not be mixing your
 legacy data and Unicode in the same string.
 
+This pragma also affects encoding of the 0x80..0xFF code point range:
+normally characters in that range are left as eight-bit bytes (unless
+they are combined with characters with code points 0x100 or larger,
+in which case all characters need to become UTF-8 encoded), but if
+the C<encoding> pragma is present, even the 0x80..0xFF range always
+gets UTF-8 encoded.
+
 If no encoding is specified, the environment variable L<PERL_ENCODING>
 is consulted.  If that fails, "latin1" (ISO 8859-1) is assumed.  If no
 encoding can be found, C<Unknown encoding '...'> error will be thrown.
 
+Note if you want to get back to the original byte encoding, you need
+to use things like I/O with encoding discplines (see L<open>) or the
+Encode module, since C<no encoding> (or re-C<encoding>) do not work.
+
 =head1 KNOWN PROBLEMS
 
 For native multibyte encodings (either fixed or variable length)
 the current implementation of the regular expressions may introduce
 recoding errors for longer regular expression literals than 127 bytes.
 
+The encoding pragma is not supported on EBCDIC platforms.
+(Porters wanted.)
+
 =head1 SEE ALSO
 
 L<perlunicode>, L<Encode>