1package Encode::Unicode; 2 3use strict; 4use warnings; 5no warnings 'redefine'; 6 7our $VERSION = do { my @r = ( q$Revision: 2.7 $ =~ /\d+/g ); sprintf "%d." . "%02d" x $#r, @r }; 8 9use XSLoader; 10XSLoader::load( __PACKAGE__, $VERSION ); 11 12# 13# Object Generator 8 transcoders all at once! 14# 15 16require Encode; 17 18our %BOM_Unknown = map { $_ => 1 } qw(UTF-16 UTF-32); 19 20for my $name ( 21 qw(UTF-16 UTF-16BE UTF-16LE 22 UTF-32 UTF-32BE UTF-32LE 23 UCS-2BE UCS-2LE) 24 ) 25{ 26 my ( $size, $endian, $ucs2, $mask ); 27 $name =~ /^(\w+)-(\d+)(\w*)$/o; 28 if ( $ucs2 = ( $1 eq 'UCS' ) ) { 29 $size = 2; 30 } 31 else { 32 $size = $2 / 8; 33 } 34 $endian = ( $3 eq 'BE' ) ? 'n' : ( $3 eq 'LE' ) ? 'v' : ''; 35 $size == 4 and $endian = uc($endian); 36 37 $Encode::Encoding{$name} = bless { 38 Name => $name, 39 size => $size, 40 endian => $endian, 41 ucs2 => $ucs2, 42 } => __PACKAGE__; 43} 44 45use base qw(Encode::Encoding); 46 47sub renew { 48 my $self = shift; 49 $BOM_Unknown{ $self->name } or return $self; 50 my $clone = bless {%$self} => ref($self); 51 $clone->{renewed}++; # so the caller knows it is renewed. 52 return $clone; 53} 54 55# There used to be a perl implemntation of (en|de)code but with 56# XS version is ripe, perl version is zapped for optimal speed 57 58*decode = \&decode_xs; 59*encode = \&encode_xs; 60 611; 62__END__ 63 64=head1 NAME 65 66Encode::Unicode -- Various Unicode Transformation Formats 67 68=cut 69 70=head1 SYNOPSIS 71 72 use Encode qw/encode decode/; 73 $ucs2 = encode("UCS-2BE", $utf8); 74 $utf8 = decode("UCS-2BE", $ucs2); 75 76=head1 ABSTRACT 77 78This module implements all Character Encoding Schemes of Unicode that 79are officially documented by Unicode Consortium (except, of course, 80for UTF-8, which is a native format in perl). 81 82=over 4 83 84=item L<http://www.unicode.org/glossary/> says: 85 86I<Character Encoding Scheme> A character encoding form plus byte 87serialization. There are Seven character encoding schemes in Unicode: 88UTF-8, UTF-16, UTF-16BE, UTF-16LE, UTF-32 (UCS-4), UTF-32BE (UCS-4BE) and 89UTF-32LE (UCS-4LE), and UTF-7. 90 91Since UTF-7 is a 7-bit (re)encoded version of UTF-16BE, It is not part of 92Unicode's Character Encoding Scheme. It is separately implemented in 93Encode::Unicode::UTF7. For details see L<Encode::Unicode::UTF7>. 94 95=item Quick Reference 96 97 Decodes from ord(N) Encodes chr(N) to... 98 octet/char BOM S.P d800-dfff ord > 0xffff \x{1abcd} == 99 ---------------+-----------------+------------------------------ 100 UCS-2BE 2 N N is bogus Not Available 101 UCS-2LE 2 N N bogus Not Available 102 UTF-16 2/4 Y Y is S.P S.P BE/LE 103 UTF-16BE 2/4 N Y S.P S.P 0xd82a,0xdfcd 104 UTF-16LE 2/4 N Y S.P S.P 0x2ad8,0xcddf 105 UTF-32 4 Y - is bogus As is BE/LE 106 UTF-32BE 4 N - bogus As is 0x0001abcd 107 UTF-32LE 4 N - bogus As is 0xcdab0100 108 UTF-8 1-4 - - bogus >= 4 octets \xf0\x9a\af\8d 109 ---------------+-----------------+------------------------------ 110 111=back 112 113=head1 Size, Endianness, and BOM 114 115You can categorize these CES by 3 criteria: size of each character, 116endianness, and Byte Order Mark. 117 118=head2 by size 119 120UCS-2 is a fixed-length encoding with each character taking 16 bits. 121It B<does not> support I<surrogate pairs>. When a surrogate pair 122is encountered during decode(), its place is filled with \x{FFFD} 123if I<CHECK> is 0, or the routine croaks if I<CHECK> is 1. When a 124character whose ord value is larger than 0xFFFF is encountered, 125its place is filled with \x{FFFD} if I<CHECK> is 0, or the routine 126croaks if I<CHECK> is 1. 127 128UTF-16 is almost the same as UCS-2 but it supports I<surrogate pairs>. 129When it encounters a high surrogate (0xD800-0xDBFF), it fetches the 130following low surrogate (0xDC00-0xDFFF) and C<desurrogate>s them to 131form a character. Bogus surrogates result in death. When \x{10000} 132or above is encountered during encode(), it C<ensurrogate>s them and 133pushes the surrogate pair to the output stream. 134 135UTF-32 (UCS-4) is a fixed-length encoding with each character taking 32 bits. 136Since it is 32-bit, there is no need for I<surrogate pairs>. 137 138=head2 by endianness 139 140The first (and now failed) goal of Unicode was to map all character 141repertoires into a fixed-length integer so that programmers are happy. 142Since each character is either a I<short> or I<long> in C, you have to 143pay attention to the endianness of each platform when you pass data 144to one another. 145 146Anything marked as BE is Big Endian (or network byte order) and LE is 147Little Endian (aka VAX byte order). For anything not marked either 148BE or LE, a character called Byte Order Mark (BOM) indicating the 149endianness is prepended to the string. 150 151CAVEAT: Though BOM in utf8 (\xEF\xBB\xBF) is valid, it is meaningless 152and as of this writing Encode suite just leave it as is (\x{FeFF}). 153 154=over 4 155 156=item BOM as integer when fetched in network byte order 157 158 16 32 bits/char 159 ------------------------- 160 BE 0xFeFF 0x0000FeFF 161 LE 0xFFFe 0xFFFe0000 162 ------------------------- 163 164=back 165 166This modules handles the BOM as follows. 167 168=over 4 169 170=item * 171 172When BE or LE is explicitly stated as the name of encoding, BOM is 173simply treated as a normal character (ZERO WIDTH NO-BREAK SPACE). 174 175=item * 176 177When BE or LE is omitted during decode(), it checks if BOM is at the 178beginning of the string; if one is found, the endianness is set to 179what the BOM says. If no BOM is found, the routine dies. 180 181=item * 182 183When BE or LE is omitted during encode(), it returns a BE-encoded 184string with BOM prepended. So when you want to encode a whole text 185file, make sure you encode() the whole text at once, not line by line 186or each line, not file, will have a BOM prepended. 187 188=item * 189 190C<UCS-2> is an exception. Unlike others, this is an alias of UCS-2BE. 191UCS-2 is already registered by IANA and others that way. 192 193=back 194 195=head1 Surrogate Pairs 196 197To say the least, surrogate pairs were the biggest mistake of the 198Unicode Consortium. But according to the late Douglas Adams in I<The 199Hitchhiker's Guide to the Galaxy> Trilogy, C<In the beginning the 200Universe was created. This has made a lot of people very angry and 201been widely regarded as a bad move>. Their mistake was not of this 202magnitude so let's forgive them. 203 204(I don't dare make any comparison with Unicode Consortium and the 205Vogons here ;) Or, comparing Encode to Babel Fish is completely 206appropriate -- if you can only stick this into your ear :) 207 208Surrogate pairs were born when the Unicode Consortium finally 209admitted that 16 bits were not big enough to hold all the world's 210character repertoires. But they already made UCS-2 16-bit. What 211do we do? 212 213Back then, the range 0xD800-0xDFFF was not allocated. Let's split 214that range in half and use the first half to represent the C<upper 215half of a character> and the second half to represent the C<lower 216half of a character>. That way, you can represent 1024 * 1024 = 2171048576 more characters. Now we can store character ranges up to 218\x{10ffff} even with 16-bit encodings. This pair of half-character is 219now called a I<surrogate pair> and UTF-16 is the name of the encoding 220that embraces them. 221 222Here is a formula to ensurrogate a Unicode character \x{10000} and 223above; 224 225 $hi = ($uni - 0x10000) / 0x400 + 0xD800; 226 $lo = ($uni - 0x10000) % 0x400 + 0xDC00; 227 228And to desurrogate; 229 230 $uni = 0x10000 + ($hi - 0xD800) * 0x400 + ($lo - 0xDC00); 231 232Note this move has made \x{D800}-\x{DFFF} into a forbidden zone but 233perl does not prohibit the use of characters within this range. To perl, 234every one of \x{0000_0000} up to \x{ffff_ffff} (*) is I<a character>. 235 236 (*) or \x{ffff_ffff_ffff_ffff} if your perl is compiled with 64-bit 237 integer support! 238 239=head1 Error Checking 240 241Unlike most encodings which accept various ways to handle errors, 242Unicode encodings simply croaks. 243 244 % perl -MEncode -e'$_ = "\xfe\xff\xd8\xd9\xda\xdb\0\n"' \ 245 -e'Encode::from_to($_, "utf16","shift_jis", 0); print' 246 UTF-16:Malformed LO surrogate d8d9 at /path/to/Encode.pm line 184. 247 % perl -MEncode -e'$a = "BOM missing"' \ 248 -e' Encode::from_to($a, "utf16", "shift_jis", 0); print' 249 UTF-16:Unrecognised BOM 424f at /path/to/Encode.pm line 184. 250 251Unlike other encodings where mappings are not one-to-one against 252Unicode, UTFs are supposed to map 100% against one another. So Encode 253is more strict on UTFs. 254 255Consider that "division by zero" of Encode :) 256 257=head1 SEE ALSO 258 259L<Encode>, L<Encode::Unicode::UTF7>, L<http://www.unicode.org/glossary/>, 260L<http://www.unicode.org/unicode/faq/utf_bom.html>, 261 262RFC 2781 L<http://www.ietf.org/rfc/rfc2781.txt>, 263 264The whole Unicode standard L<http://www.unicode.org/unicode/uni2book/u2.html> 265 266Ch. 15, pp. 403 of C<Programming Perl (3rd Edition)> 267by Larry Wall, Tom Christiansen, Jon Orwant; 268O'Reilly & Associates; ISBN 0-596-00027-8 269 270=cut 271