Muntilingualizaion of w3m 
                                                              2003/03/08
                                                              H. Sakamoto

Introduction

  I have tried the muntilingualization of w3m (w3m-m17n).
  The patch for w3m-0.4.1 is available on the following site.

    http://www2u.biglobe.ne.jp/~hsaka/w3m/index.html#m17n
                                          patch/w3m-0.4.1-m17n-20030308.tar.gz
                                          patch/README.m17n

  It is a development version. And enough test is not preformed because
  I can understand Japanese only. Please use, test, and report bugs.

  Now, w3m-m17n has following functions.

Supported encoding schemes (character set)

  * Japanese
      EUC-JP           - US_ASCII, JIS X 0208, JIS X 0201, JIS X 0212
      (EUC-JISX0213)     (JIS X 0213)
      ISO-2022-JP      - US_ASCII, JIS X 0208, JIS X 0201, JIS X 0212, etc.
      ISO-2022-JP-2    - US_ASCII, JIS X 0208, JIS X 0201, JIS X 0212,
                         GB 2312, KS X 1001, ISO 8859-1, ISO 8859-7, etc.
      ISO-2022-JP-3    - US_ASCII, JIS X 0208, JIS X 0201, JIS X 0213, etc.
      Shift_JIS(CP932) - US_ASCII, JIS X 0208, JIS X 0201, CP932 extension
      Shift_JISX0213   - US_ASCII, JIS X 0208, JIS X 0201, JIS X 0213
  * Chinese (simplified)
      EUC-CN(GB2312) - US_ASCII, GB 2312
      ISO-2022-CN    - US_ASCII, GB 2312, CNS-11643-1,..7, etc.
      GBK(CP936)     - US_ASCII, GB 2312, GBK
      GB18030        - US_ASCII, GB 2312, GBK, GB18030, Unicode,
      HZ-GB-2312     - US_ASCII, GB 2312
  * Chinese (Taiwan, tradisional)
      EUC-TW        - US_ASCII, CNS 11643-1,..16
      ISO-2022-CN   - US_ASCII, CNS-11643-1,..7, GB 2312, etc.
      Big5          - Big5
      HKSCS         - Big5, HKSCS
  * Korean
      EUC-KR        - US_ASCII, KS X 1001 Wansung
      ISO-2022-KR   - US_ASCII, KS X 1001 Wansung, etc.
      Johab         - US_ASCII, KS X 1001 Johab
      UHC(CP949)    - US_ASCII, KS X 1001 Wansung, UHC
  * Vietnamese
      TCVN-5712 VN-1, VISCII 1.1, VPS, CP1258
  * Thai
      TIS-620 (ISO-8859-11), CP874
  * Other
      US_ASCII, ISO-8859-1 - 10, 13 - 15,
      KOI8-R, KOI8-U, NeXT, CP437, CP737, CP775, CP850, CP852, CP855, CP856,
      CP857, CP860, CP861, CP862, CP863, CP864, CP865, CP866, CP869, CP1006,
      CP1250, CP1251, CP1252, CP1253, CP1254, CP1255, CP1256, CP1257
  * Unicode (UCS-4)
      UTF-8, UTF-7

  NOTE:
    * The left part of JIS X 0201 and GB 1988 (Chinese ASCII) are
      treated as US_ASCII because they are used in tags of HTML document.
      Another variant of US_ASCII is treated without change.
    * JIS C 6226(old JIS) is treated as JIS X 0208.
    * The sequence '~\n' of HZ is not supported.

Display

  There are two method for multilingual diplay.

  (1) kterm + ISO-2022-JP/CN/KR

    * kterm can handle JIS X 0213, CNS 11643, if the following patch
      is applied.
        http://www.st.rim.or.jp/~hanataka/kterm-6.2.0.ext02.patch.gz

    * Specify the fontList for kterm with -fl option or in ~/.Xdefaults.
    
        -fl "*--16-*-jisx0213.2000-*,\
             *--16-*-jisx0212.1990-0,\
             *--16-*-ksc5601.1987-0,\
             *--16-*-gb2312.1980-0,\
             *--16-*-cns11643.1992-*,\
             *--16-*-iso8859-*"

      Fonts of JIS X 0213 exist in
        http://www.mars.sphere.ne.jp/imamura/jisx0213.html

    * Set the "display_charset" to ISO-2022-JP(or ISO-2022-JP-2, KR, CN),
      and "strict_iso2022" to OFF on the option pannel. (see below)

  (2) xterm + UTF-8

    * Use xterm (xterm-140 or later) of XFree86.
        http://www.clark.net/pub/dickey/xterm/xterm.html

    * Fonts of Unicode exist in
        http://www.cl.cam.ac.uk/~mgk25/ucs-fonts.html
        http://openlab.ring.gr.jp/efont/index.html.en

    * Use xterm with -u8 option.
      The fonts are specified such as
        -fn "*-medium-*--13-*-iso10646-1" \
        -fb "*-bold-*--13-*-iso10646-1" \
        -fw "*-medium-*-ja-13-*-iso10646-1"

    * Set the "display_charset" to UTF-8.
      And, it is better that "pre_conv" is ON.

  (3) mlterm + ISO-2022-JP/KR/CN

    * Homepage
        http://mlterm.sourceforge.net/

    * Set encoding of mlterm to ISO-2022-JP/KR/CN or UTF-8.

    * Set the "display_charset" to ISO-2022-JP/KR/CN or UTF-8.

Command line options

   -I <document charset>
   -O <display/output charset>

        j(p):      ISO-2022-JP
        j(p)2:     ISO-2022-JP-2
        j(p)3:     ISO-2022-JP-3
        cn:        ISO-2022-CN
        kr:        ISO-2022-KR
        e(j):      EUC-JP
        ec,g(b):   EUC-CN(GB2312)
        et:        EUC-TW
        ek:        EUC-KR
        s(jis):    Shift_JIS
        sjisx0213: Shift_JISX0213
        gbk:       GBK
        gb18030:   GB18030
        h(z):      HZ-GB-2312
        b(ig5):    Big5
        hk(scs):   HKSCS
        jo(hab):   Johab
        uhc:       UHC
        l?:        ISO-8859-?
        t(is):     TIS-620(ISO-8859-11)
        tc(vn):    TCVN-5712 VN-1
        v(iscii):  VISCII 1.1
        vp(s):     VPS
        ko(i8r):   KOI8-R
        koi8u:     KOI8-U
        n(ext):    NeXT
        cp???:     CP???
        w12??:     CP12??
        u(tf8):    UTF-8
        u(tf)7:    UTF-7

Option pannel

   display_charset
       Display charset.
   document_charset
       Defalut Document charset.
   auto_detect
       Automatic charset detect when loading. (Default: ON)
   system_charset
       System charset. It is used for configuration files and file name.
   follow_locale
       System charset follows locale($LANG). (Default: ON)
   ext_halfdump
       Output with display charset when -halfdump.
   search_conv
       Adjust search string for document charset. (Default: ON)
   use_wide
       Use multi column characters. (Default: ON)
   use_combining
       Use combining characters. (Default: ON)
   use_language_tag
       Use Unicode language tags. (Default: ON)
   ucs_conv
       Charset conversion using Unicode map. (Default: ON)
   pre_conv
       Charset conversion when loading. (Default: OFF)
   fix_width
       Fix character width when conversion. (Default: ON)
       If it is OFF, the rendering may collapse.
   use_gb12345_map
       Use GB 12345 Unicode map instead of GB 2312's. (Default: OFF)
       If it is ON, GB2312 can be converted to Big5, EUC-TW, or EUC-JP.
   use_jisx0201
       Use JIS X 0201 Roman for ISO-2022-JP. (Default: OFF)
   use_jisc6226
       Use JIS C 6226:1978 for ISO-2022-JP. (Default: OFF)
   use_jisx0201k
       Use JIS X 0201 Katakana. (Default: OFF)
   use_jisx0212
       Use JIS X 0212:1990 (Supplemental Kanji). (Default: OFF)
   use_jisx0213
       Use JIS X 0213:2000 (2000JIS). (Default: OFF)
   strict_iso2022
       Strict ISO-2022-JP/KR/CN. (Default: ON)
       If it is OFF, all ISO 2022 base character set can be displayed
       with ISO-2022-JP/KR/CN.
   east_asian_width
       Use double width for some Unicode characters. (Default: OFF)
       If it is ON, treat East Asian Ambiguous characters as double width.
   gb18030_as_ucs
       Treat 4 bytes char. of GB18030 as Unicode. (Default: OFF)
   simple_preserve_space
       Simple Preserve space.
       If it is ON, a space is remained in Japanese and some other languages.

   alt_entity
       Use alternate expression with ASCII for entities. (Default: ON)
       If it is OFF, entities are treated as ISO 8859-1
   graphic_char
       Use DEC special graphics for border of table and menu.
       If it is OFF, ruled line is used with CJK charset or UTF-8.

Code conversion

  The following special code conversions are supported.
    * EUC-JP <-> ISO-2022-JP <-> Shift-JIS
    * EUC-CN <-> ISO-2022-CN <-> HZ-GB-2312
    * EUC-TW <-> ISO-2022-CN
    * EUC-KR <-> ISO-2022-KR <-> Johab (only Symbol and Hanja)

  Other conversions are based on Unicode.

Change document charset

   Press '=' (show document infomation), and select document charaset.

   If you specify the following keymaps,
     keymap C CHARSET
     keymap M-c DEFAULT_CHARSET
   you can press `C' to change the current document charset,
   and `M-c' to change the default document charset.

Line Editing 

  Input coding system is followed by display coding system.

  NOTE:
    * HZ can not be used as input coding system.
    * Input with ISO-2022-CN or ISO-2022-KR is perhaps failure, because
      SI(\017) and SO(\016) are already assigned as other command key.
      (SO is assigned as `next-history'). If you want to use SI and SO,
      press C-@(^@). After that, SI, SO, SS2, SS3, LS2, and LS3 of
      7bit ISO-2022 are recognited. When you press C-@ again, the default
      binding is set.

Regular expression

   Multilingual regular expression is supported.

-----------------------------------
Change log

2003/03/08      w3m-0.4.1-m17n-20030308
 * Base on w3m-0.4.1

2003/02/24      w3m-0.4-m17n-20030224
 * Base on w3m-0.4

2003/02/11      w3m-0.4rc1-m17n-20030211
 * Base on w3m-0.4rc1

2003/02/07      w3m-0.3.2.2-m17n-20030207
 * Base on w3m-0.3.2.2+cvs-1.742

2003/02/01      w3m-0.3.2.2-m17n-20030201
 * Base on w3m-0.3.2.2+cvs-1.734

2003/01/31      w3m-0.3.2.2-m17n-20030131
 * Base on w3m-0.3.2.2+cvs-1.732

2003/01/23      w3m-0.3.2.2-m17n-20030123
 * Base on w3m-0.3.2.2+cvs-1.705

2003/01/22      w3m-0.3.2.2-m17n-20030122
 * Base on w3m-0.3.2.2+cvs-1.699

2003/01/01      w3m-0.3.2.2-m17n-20030101
 * Base on w3m-0.3.2.2+cvs-1.655

2002/12/22      w3m-0.3.2.2-m17n-20021222
 * Base on w3m-0.3.2.2+cvs-1.640

2002/12/19      w3m-0.3.2.2-m17n-20021219
 * Base on w3m-0.3.2.2+cvs-1.635

2002/12/07      w3m-0.3.2.2-m17n-20021207
 * Base on w3m-0.3.2.2+cvs-1.599
 * Fixed a problem on int != long system

2002/11/27	w3m-0.3.2.1-m17n-20021127
 * Base on w3m-0.3.2.1+cvs-1.562

2002/11/20	w3m-0.3.2-m17n-20021120
 * Base on w3m-0.3.2+cvs-1.538

2002/11/18
 * Added UTF-7 to auto detection of charset.

2002/11/16	w3m-0.3.2-m17n-20021116
 * Base on w3m-0.3.2+cvs-1.526

2002/11/13	w3m-0.3.2-m17n-20021113
 * Base on w3m-0.3.2+cvs-1.506

2002/11/12	w3m-0.3.2-m17n-20021112
 * Base on w3m-0.3.2+cvs-1.498

2002/11/09	w3m-0.3.2-m17n-20021109
 * Base on w3m-0.3.2+cvs-1.490

2002/11/07	w3m-0.3.2-m17n-20021107
 * Base on w3m-0.3.2
 * Applied [w3m-dev 03371]

2002/10/22	w3m-0.3.1-m17n-20021022
 * Base on w3m-0.3.1+cvs-1.444

2002/07/17	w3m-0.3.1-m17n-20020717
 * Base on w3m-0.3.1

2002/05/29	w3m-0.3-m17n-20020529
 * Base on w3m-0.3+cvs-1.379.

2002/03/16	w3m-0.3-m17n-20020316
 * Base on w3m-0.3+cvs-1.353.

2002/03/11	w3m-0.3-m17n-20020311
 * Base on w3m-0.3+cvs-1.342.
 * Some bug fixes.

2002/02/16	w3m-0.2.5-m17n-20020216
 * Base on w3m-0.2.5+cvs-1.319.
 * Added an option "use_wide"

2002/02/05	w3m-0.2.5-m17n-20020205
 * Base on w3m-0.2.5+cvs-1.302.

2002/02/02	w3m-0.2.5-m17n-20020202
 * Base on w3m-0.2.5+cvs-1.291.

2002/01/31	w3m-0.2.4-m17n-20020131
 * Base on w3m-0.2.4+cvs-1.278.

2002/01/29	w3m-0.2.4-m17n-20020129
 * Base on w3m-0.2.4+cvs-1.268.
 * Some bug fixes.

2002/01/28	w3m-0.2.4-m17n-20020128
 * Base on w3m-0.2.4+cvs-1.265.

2002/01/08	w3m-0.2.4-m17n-20020108
 * Base on w3m-0.2.4.

2002/01/07
 * Replaced some wc_conv,wc_Str_conv with wc_conv_strict,wc_Str_conv_strict.

2001/12/31
 * Added the conversion between HKSCS and Unicode.
 * Changed the conversion table between Big5 and Unicode.
 * Deleted the special conversion between Big5 and CNS11643.
 * Fixed HKSCS.

2001/12/30	w3m-0.2.3.2-m17n-20011230
 * Base on w3m-0.2.3.2+cvs-1.196.

2001/12/22	w3m-0.2.3.2-m17n-20011222
 * Base on w3m-0.2.3.2.
 * [w3m-dev-en 00660] can't compile if INET6 is defined
 * [w3m-dev-en 00663] double meanings for WC_N_??? 

2001/12/21	w3m-0.2.3.1-m17n-20011221
 * Base on w3m-0.2.3.1.
 * Support of HKSCS, KOI8-U, UTF-7.
   The conversion table between HKSCS and Unicode is not yet available.
 * Add the conversion between ISO 8859-16 and Unicode.
 * Add option 'ext_halfdump'.

2001/04/14	w3m-(0.2.1)-m17n-0.20
 * Support of UTF-7.
 * [w3m-dev 01913] ([w3m-dev-en 00452])

2001/04/12	w3m-(0.2.1)-m17n-0.19
 * TILDE of JISX0212, JISX0213 -> FULLWIDTH TILDE of Unicode.
 * MICRO SIGN of Unicode -> GREEK SMALL MU of JISX0208.
 * [w3m-dev 01892], [w3m-dev 01894], [w3m-dev 01898], [w3m-dev 01902]

2001/03/31
 * Changed implement of <_SYMBOL> again.
 * When -dump option, "pre_conv" is false as default.

2001/03/29
 * Support combining characters of TCVN 5712.
 * [w3m-dev 01873], [w3m-dev-en 00411].

2001/03/28
 * Setting -suffix="" can be okay in confiugre. (thanks to naddy!)
 * Bugfix: when #define USE_SSL and #undef USE_SSL_VERIFY, rc.c
   doesn't compile. (thanks to naddy!)
 * [w3m-dev 01859].
 * Bugfix: 0xA0 is error in Shift-JIS.
 * Changed implement of <_SYMBOL> ([w3m-dev 01852]).

2001/03/24	w3m-(0.2.1)-m17n-0.18
 * Base on w3m-0.2.1.
 * [w3m-dev 01703], [w3m-dev 01814], [w3m-dev 01823]
 * Separated ISO-2022-JP-3 from ISO-2022-JP.
 * Improved auto detection.

2001/03/23
 * Base on w3m-0.2.0.

2001/03/21
 * Added functions (CHARSET and DEFAULT_CHARSET).
 * Improved document charset detection of frame HTML.

2001/03/20
 * Conversion from FULL WIDTH variant except ASCII to normal character.

2001/03/18	w3m-(0.1.11-pre-hsaka24)-m17n-0.17
 * Based on "[w3m-dev 01779] w3m-0.1.11-pre-hsaka24".
 * Prefer JIS X 0213 than JIS X 0212.

2001/03/14      w3m-(0.1.11-pre-kokb23)-m17n-0.16
 * Add the conversion between JIS X 0213 and Unicode Extention B.
 * Bugfix: conversion between JIS X 0213 and Unicode.
 * Bugfix: treat UHC as Hangul.
 * Ignore "search_conv" if "pre_conv" is ON.

2001/03/09	w3m-(0.1.11-pre-kokb23)-m17n-0.15
 * Improvement of wc_wchar_t (mainly for Unicode).
 * Some bugfixes for Unicode.
 * Ignore "use_gb12345_map" option when output with GBK or GB18030.
 * When -dump option, "prev_conv" is always true.
 * when -dump or -halfdump option, some proccessing is skiped.
 * Get system charset from the environment variable LC_CTYPE -> LANG -> LC_ALL.
 * Bugfixes: [w3m-dev 01724], [w3m-dev 01726], [w3m-dev 01752],
   [w3m-dev 01753], [w3m-dev 01754]

2001/03/06	w3m-(0.1.11-pre-kokb23)-m17n-0.14
 * Support of Language tag (UTR#7).
 * Bugfix: conversion between GB18030, Johab and Unicode.

2001/03/04	w3m-(0.1.11-pre-kokb23)-m17n-0.13
 * Support of GBK(CP936), GB18030, UHC(CP949) !
 * Unicode mapping table of GB2312 and GB12345 became compatible with
   CP936, GB18030. (Code point: 0xA1A4, 0xA1AA)
 * Allow 0xFFFE and 0xFFFF in Uncide (due to compatibility with GB18030).
 * Bugfix: code point of NBSP in Unicode.

2001/03/03	w3m-(0.1.11-pre-kokb23)-m17n-0.12
 * I wrote English README.m17n.

-------------------------------------------
Hironori Sakamoto <hsaka@mth.biglobe.ne.jp>
 http://www2u.biglobe.ne.jp/~hsaka/