Encode::HanExtra3pm

Usage

    use Encode;
    # Traditional Chinese
    $euc_tw = encode("euc-tw", $utf8);   # loads Encode::HanExtra implicitly
    $utf8   = decode("euc-tw", $euc_tw); # ditto
    # Simplified Chinese
    $gb18030 = encode("gb18030", $utf8);    # loads Encode::HanExtra implicitly
    $utf8    = decode("gb18030", $gb18030); # ditto

Version

This document describes version 0.23 of Encode::HanExtra, released November 10, 2007.

Overview

Perl 5.7.3 and later ships with an adequate set of Chinese encodings, including the commonly used "CP950", "CP936" (also known as "GBK"), "Big5" (alias for "Big5-Eten"), "Big5-HKSCS", "EUC-CN", "HZ", and "ISO-IR-165".

However, the numbers of Chinese encodings are staggering, and a complete coverage will easily increase the size of perl distribution by several megabytes; hence, this CPAN module tries to provide the rest of them.

If you are using Perl 5.8 or later, Encode::CN and Encode::TW will automatically load the extra encodings for you, so there's no need to explicitly write "use Encode::HanExtra" if you are using one of them already.

Encodings

This version includes the following encoding tables:

  Canonical   Alias                             Description
  -----------------------------------------------------------------------------
  big5-1984   /\b(tca-)?big5-?(19)?84$/i        TCA's original Big5-1984
  big5-2003   /\b(cmex-)?big5-?(20)?03$/i       Big5-2003 (national standard)
  big5ext     /\b(cmex-)?big5-?e(xt)?$/i        CMEX's Big5e Extension
  big5plus    /\b(cmex-)?big5-?p(lus)?$/i       CMEX's Big5+ Extension
              /\b(cmex-)?big5\+$/i
  cccii       /\b(ccag-)?cccii$/i               Chinese Character Code for
                                                Information Interchange
  cns11643-1  /\bCNS[-_ ]?11643[-_]1$/i         Taiwan's CNS map, plane 1
  cns11643-2  /\bCNS[-_ ]?11643[-_]2$/i         Taiwan's CNS map, plane 2
  cns11643-3  /\bCNS[-_ ]?11643[-_]3$/i         Taiwan's CNS map, plane 3
  cns11643-4  /\bCNS[-_ ]?11643[-_]4$/i         Taiwan's CNS map, plane 4
  cns11643-5  /\bCNS[-_ ]?11643[-_]5$/i         Taiwan's CNS map, plane 5
  cns11643-6  /\bCNS[-_ ]?11643[-_]6$/i         Taiwan's CNS map, plane 6
  cns11643-7  /\bCNS[-_ ]?11643[-_]7$/i         Taiwan's CNS map, plane 7
  cns11643-f  /\bCNS[-_ ]?11643[-_]f$/i         Taiwan's CNS map, plane F
  euc-tw      /\beuc.*tw$/i                     EUC (Extended Unix Character)
              /\btw.*euc$/i
  gb18030     /\bGB[-_ ]?18030$/i               GBK with Traditional Characters
  unisys      /\bunisys$/i                      Unisys Traditional Chinese
  unisys-sosi1                                  Unisys SOSI1 transport encoding
  unisys-sosi2                                  Unisys SOSI2 transport encoding

Detailed descriptions are as follows:

BIG5-1984

This is the original Big5 encoding made by TCA Taiwan.

BIG5-2003

This revised encoding is now national standard, as an appendix of CNS11643.

BIG5PLUS

This encoding, while not heavily used, is an attempt to bring all Taiwan's conflicting internal-use encodings together, and fit it as an extension to the widely-deployed Big5 range, by CMEX Taiwan.

BIG5EXT

The CMEX's second (and less ambitious) try at unifying the most commonly used characters not covered by Big5, while not polluting out of the 94x94 arragement like BIG5PLUS did.

CCCII

The earliest (and most sophisticated) Traditional Chinese encoding, with a three-byte raw character map, made in 1980 by the Chinese Character Analysis Group (CCAG), used mostly in library systems.

EUC-TW

The EUC transport version of "CNS11643" (planes 1-7), the comprehensive character set used by the Taiwan government.

CNS11643-*

The raw character map extracted from the Unihan database, including the plane F which wasn't included in "EUC-TW".

GB18030

An extension to GBK, this encoding lists most Han characters (both simplified and traditional), as well as some other encodings used by other peoples in China.

UNISYS

Unisys System's internal Chinese mapping.

Notes

If you are looking for ways to transliterate between Simplified and Traditional Chinese, please take a look at Encode::HanConvert. Note that the direct mapping via Unicode is lossy, and usually doesn't work at all.

Please send me suggestions if you want to see more encoding added, such as "BIG5-GCCS" (superseded by "BIG5-HKSCS"). Other suggestions are welcome, too.

Acknowledgements

Some of the maps here are generated from GNU libiconv's test files, with kind permission from Bruno Haible.

Map for "BIG5PLUS" is generated from the BIG52UCS.TXT file, courtesy of CMEX Taiwan (Chinese Microcomputer Extended Foundation, <http://www.cmex.org.tw/>).

Map for "BIG5-1984" is supplied by imacat.

Map for "CCCII" is supplied by the Koha Taiwan project.