Update localcharset.c and config.charset from libiconv.

6 files changed, 460 insertions, 52 deletions

diff --git a/intl/ChangeLog b/intl/ChangeLog
index 8834cc3..a562850 100644
--- a/intl/ChangeLog
+++ b/intl/ChangeLog
@@ -1,3 +1,8 @@
+2001-03-09  Bruno Haible  <haible@clisp.cons.org>
+
+	* config.charset: Update from libiconv-1.6.
+	* localcharset.c: Likewise.
+
 2001-02-25  Ulrich Drepper  <drepper@redhat.com>
 
 	* locale.alias: Don't use nb_NO but define aliases for it.
diff --git a/intl/config.charset b/intl/config.charset
index bf21d99..0844094 100755
--- a/intl/config.charset
+++ b/intl/config.charset
@@ -33,27 +33,37 @@
 #       name                         used by which systems         a MIME name?
 #   ASCII, ANSI_X3.4-1968     glibc solaris freebsd
 #   ISO-8859-1                glibc aix hpux irix osf solaris freebsd   yes
-#   ISO-8859-2                glibc aix hpux irix solaris freebsd       yes
+#   ISO-8859-2                glibc aix hpux irix osf solaris freebsd   yes
 #   ISO-8859-3                glibc                                     yes
-#   ISO-8859-4                solaris freebsd                           yes
-#   ISO-8859-5                glibc aix hpux irix solaris freebsd       yes
+#   ISO-8859-4                osf solaris freebsd                       yes
+#   ISO-8859-5                glibc aix hpux irix osf solaris freebsd   yes
 #   ISO-8859-6                glibc aix hpux solaris                    yes
 #   ISO-8859-7                glibc aix hpux irix osf solaris           yes
-#   ISO-8859-8                glibc aix hpux solaris                    yes
+#   ISO-8859-8                glibc aix hpux osf solaris                yes
 #   ISO-8859-9                glibc aix hpux irix osf solaris           yes
 #   ISO-8859-13               glibc
-#   ISO-8859-15               glibc aix solaris freebsd
+#   ISO-8859-15               glibc aix osf solaris freebsd
 #   KOI8-R                    glibc solaris freebsd                     yes
 #   KOI8-U                    glibc freebsd                             yes
-#   CP850                     aix osf
+#   CP437                     dos
+#   CP775                     dos
+#   CP850                     aix osf dos
+#   CP852                     dos
+#   CP855                     dos
 #   CP856                     aix
-#   CP866                     freebsd
-#   CP874                     win32
+#   CP857                     dos
+#   CP861                     dos
+#   CP862                     dos
+#   CP864                     dos
+#   CP865                     dos
+#   CP866                     freebsd dos
+#   CP869                     dos
+#   CP874                     win32 dos
 #   CP922                     aix
-#   CP932                     aix win32
+#   CP932                     aix win32 dos
 #   CP943                     aix
-#   CP949                     win32
-#   CP950                     win32
+#   CP949                     osf win32 dos
+#   CP950                     win32 dos
 #   CP1046                    aix
 #   CP1124                    aix
 #   CP1129                    aix
@@ -66,16 +76,16 @@
 #   CP1256                    win32
 #   CP1257                    win32
 #   GB2312                    glibc aix hpux irix solaris freebsd       yes
-#   EUC-JP                    glibc aix hpux irix solaris freebsd       yes
-#   EUC-KR                    glibc aix hpux irix solaris freebsd       yes
-#   EUC-TW                    glibc aix hpux irix solaris
-#   BIG5                      glibc aix hpux solaris freebsd            yes
+#   EUC-JP                    glibc aix hpux irix osf solaris freebsd   yes
+#   EUC-KR                    glibc aix hpux irix osf solaris freebsd   yes
+#   EUC-TW                    glibc aix hpux irix osf solaris
+#   BIG5                      glibc aix hpux osf solaris freebsd        yes
 #   BIG5HKSCS                 glibc
-#   GBK                       aix win32
+#   GBK                       aix osf win32 dos
 #   GB18030                   glibc
-#   SJIS                      hpux solaris freebsd
+#   SJIS                      hpux osf solaris freebsd
 #   JOHAB                     glibc win32
-#   TIS-620                   glibc aix hpux solaris
+#   TIS-620                   glibc aix hpux osf solaris
 #   VISCII                    glibc                                     yes
 #   HP-ROMAN8                 hpux
 #   HP-ARABIC8                hpux
@@ -83,7 +93,9 @@
 #   HP-HEBREW8                hpux
 #   HP-TURKISH8               hpux
 #   HP-KANA8                  hpux
-#   UTF-8                     glibc aix hpux solaris                    yes
+#   DEC-KANJI                 osf
+#   DEC-HANYU                 osf
+#   UTF-8                     glibc aix hpux osf solaris                yes
 #
 # Note: Names which are not marked as being a MIME name should not be used in
 # Internet protocols for information interchange (mail, news, etc.).
@@ -178,9 +190,28 @@ case "$os" in
 	;;
     osf*)
 	echo "ISO8859-1 ISO-8859-1"
+	echo "ISO8859-2 ISO-8859-2"
+	echo "ISO8859-4 ISO-8859-4"
+	echo "ISO8859-5 ISO-8859-5"
 	echo "ISO8859-7 ISO-8859-7"
+	echo "ISO8859-8 ISO-8859-8"
 	echo "ISO8859-9 ISO-8859-9"
+	echo "ISO8859-15 ISO-8859-15"
 	echo "cp850 CP850"
+	echo "big5 BIG5"
+	echo "dechanyu DEC-HANYU"
+	echo "dechanzi GB2312"
+	echo "deckanji DEC-KANJI"
+	echo "deckorean EUC-KR"
+	echo "eucJP EUC-JP"
+	echo "eucKR EUC-KR"
+	echo "eucTW EUC-TW"
+	echo "GBK GBK"
+	echo "KSC5601 CP949"
+	echo "sdeckanji EUC-JP"
+	echo "SJIS SJIS"
+	echo "TACTIS TIS-620"
+	echo "UTF-8 UTF-8"
 	;;
     solaris*)
 	echo "646 ASCII"
@@ -214,8 +245,8 @@ case "$os" in
 	  echo "$l.ASCII ASCII"
 	done
 	for l in da_DK de_AT de_CH de_DE en_AU en_CA en_GB en_US es_ES \
-                 fi_FI fr_BE fr_CA fr_CH fr_FR is_IS it_CH it_IT la_LN \
-                 lt_LN nl_BE nl_NL no_NO pt_PT sv_SE; do
+	         fi_FI fr_BE fr_CA fr_CH fr_FR is_IS it_CH it_IT la_LN \
+	         lt_LN nl_BE nl_NL no_NO pt_PT sv_SE; do
 	  echo "$l.ISO_8859-1 ISO-8859-1"
 	  echo "$l.DIS_8859-15 ISO-8859-15"
 	done
@@ -243,4 +274,165 @@ case "$os" in
 	# BeOS has a single locale, and it has UTF-8 encoding.
 	echo "C UTF-8"
 	;;
+    msdosdjgpp*)
+	# DJGPP 2.03 doesn't have nl_langinfo(CODESET); therefore
+	# localcharset.c falls back to using the full locale name
+	# from the environment variables.
+	echo "#"
+	echo "# The encodings given here may not all be correct."
+	echo "# If you find that the encoding given for your language and"
+	echo "# country is not the one your DOS machine actually uses, just"
+	echo "# correct it in this file, and send a mail to"
+	echo "# Juan Manuel Guerrero <st001906@hrz1.hrz.tu-darmstadt.de>"
+	echo "# and Bruno Haible <haible@clisp.cons.org>."
+	echo "#"
+	echo "C ASCII"
+	# ISO-8859-1 languages
+	echo "ca CP850"
+	echo "ca_ES CP850"
+	echo "da CP865"    # not CP850 ??
+	echo "da_DK CP865" # not CP850 ??
+	echo "de CP850"
+	echo "de_AT CP850"
+	echo "de_CH CP850"
+	echo "de_DE CP850"
+	echo "en CP850"
+	echo "en_AU CP850" # not CP437 ??
+	echo "en_CA CP850"
+	echo "en_GB CP850"
+	echo "en_NZ CP437"
+	echo "en_US CP437"
+	echo "en_ZA CP850" # not CP437 ??
+	echo "es CP850"
+	echo "es_AR CP850"
+	echo "es_BO CP850"
+	echo "es_CL CP850"
+	echo "es_CO CP850"
+	echo "es_CR CP850"
+	echo "es_CU CP850"
+	echo "es_DO CP850"
+	echo "es_EC CP850"
+	echo "es_ES CP850"
+	echo "es_GT CP850"
+	echo "es_HN CP850"
+	echo "es_MX CP850"
+	echo "es_NI CP850"
+	echo "es_PA CP850"
+	echo "es_PY CP850"
+	echo "es_PE CP850"
+	echo "es_SV CP850"
+	echo "es_UY CP850"
+	echo "es_VE CP850"
+	echo "et CP850"
+	echo "et_EE CP850"
+	echo "eu CP850"
+	echo "eu_ES CP850"
+	echo "fi CP850"
+	echo "fi_FI CP850"
+	echo "fr CP850"
+	echo "fr_BE CP850"
+	echo "fr_CA CP850"
+	echo "fr_CH CP850"
+	echo "fr_FR CP850"
+	echo "ga CP850"
+	echo "ga_IE CP850"
+	echo "gd CP850"
+	echo "gd_GB CP850"
+	echo "gl CP850"
+	echo "gl_ES CP850"
+	echo "id CP850"    # not CP437 ??
+	echo "id_ID CP850" # not CP437 ??
+	echo "is CP861"    # not CP850 ??
+	echo "is_IS CP861" # not CP850 ??
+	echo "it CP850"
+	echo "it_CH CP850"
+	echo "it_IT CP850"
+	echo "lt CP775"
+	echo "lt_LT CP775"
+	echo "lv CP775"
+	echo "lv_LV CP775"
+	echo "nb CP865"    # not CP850 ??
+	echo "nb_NO CP865" # not CP850 ??
+	echo "nl CP850"
+	echo "nl_BE CP850"
+	echo "nl_NL CP850"
+	echo "nn CP865"    # not CP850 ??
+	echo "nn_NO CP865" # not CP850 ??
+	echo "no CP865"    # not CP850 ??
+	echo "no_NO CP865" # not CP850 ??
+	echo "pt CP850"
+	echo "pt_BR CP850"
+	echo "pt_PT CP850"
+	echo "sv CP850"
+	echo "sv_SE CP850"
+	# ISO-8859-2 languages
+	echo "cs CP852"
+	echo "cs_CZ CP852"
+	echo "hr CP852"
+	echo "hr_HR CP852"
+	echo "hu CP852"
+	echo "hu_HU CP852"
+	echo "pl CP852"
+	echo "pl_PL CP852"
+	echo "ro CP852"
+	echo "ro_RO CP852"
+	echo "sk CP852"
+	echo "sk_SK CP852"
+	echo "sl CP852"
+	echo "sl_SI CP852"
+	echo "sq CP852"
+	echo "sq_AL CP852"
+	echo "sr CP852"    # CP852 or CP866 or CP855 ??
+	echo "sr_YU CP852" # CP852 or CP866 or CP855 ??
+	# ISO-8859-3 languages
+	echo "mt CP850"
+	echo "mt_MT CP850"
+	# ISO-8859-5 languages
+	echo "be CP866"
+	echo "be_BE CP866"
+	echo "bg CP866"    # not CP855 ??
+	echo "bg_BG CP866" # not CP855 ??
+	echo "mk CP866"    # not CP855 ??
+	echo "mk_MK CP866" # not CP855 ??
+	echo "ru KOI8-R"    # not CP866 ??
+	echo "ru_RU KOI8-R" # not CP866 ??
+	# ISO-8859-6 languages
+	echo "ar CP864"
+	echo "ar_AE CP864"
+	echo "ar_DZ CP864"
+	echo "ar_EG CP864"
+	echo "ar_IQ CP864"
+	echo "ar_IR CP864"
+	echo "ar_JO CP864"
+	echo "ar_KW CP864"
+	echo "ar_MA CP864"
+	echo "ar_OM CP864"
+	echo "ar_QA CP864"
+	echo "ar_SA CP864"
+	echo "ar_SY CP864"
+	# ISO-8859-7 languages
+	echo "el CP869"
+	echo "el_GR CP869"
+	# ISO-8859-8 languages
+	echo "he CP862"
+	echo "he_IL CP862"
+	# ISO-8859-9 languages
+	echo "tr CP857"
+	echo "tr_TR CP857"
+	# Japanese
+	echo "ja CP932"
+	echo "ja_JP CP932"
+	# Chinese
+	echo "zh_CN GBK"
+	echo "zh_TW CP950" # not CP938 ??
+	# Korean
+	echo "kr CP949"    # not CP934 ??
+	echo "kr_KR CP949" # not CP934 ??
+	# Thai
+	echo "th CP874"
+	echo "th_TH CP874"
+	# Other
+	echo "eo CP850"
+	echo "eo_EO CP850"
+	;;
 esac
diff --git a/intl/localcharset.c b/intl/localcharset.c
index 12214b7..201483c 100644
--- a/intl/localcharset.c
+++ b/intl/localcharset.c
@@ -1,6 +1,6 @@
 /* Determine a canonical name for the current locale's character encoding.
 
-   Copyright (C) 2000, 2001 Free Software Foundation, Inc.
+   Copyright (C) 2000-2001 Free Software Foundation, Inc.
 
    This program is free software; you can redistribute it and/or modify it
    under the terms of the GNU Library General Public License as published
@@ -194,7 +194,7 @@ get_charset_aliases ()
    into one of the canonical names listed in config.charset.
    The result must not be freed; it is statically allocated.
    If the canonical name cannot be determined, the result is a non-canonical
-   name or NULL.  */
+   name.  */
 
 #ifdef STATIC
 STATIC
@@ -214,10 +214,14 @@ locale_charset ()
 
 # else
 
-  /* On old systems which lack it, use setlocale and getenv.  */
+  /* On old systems which lack it, use setlocale or getenv.  */
   const char *locale = NULL;
 
-#  if HAVE_SETLOCALE
+  /* But most old systems don't have a complete set of locales.  Some
+     (like SunOS 4 or DJGPP) have only the C locale.  Therefore we don't
+     use setlocale here; it would return "C" when it doesn't support the
+     locale name the user has set.  */
+#  if HAVE_SETLOCALE && 0
   locale = setlocale (LC_CTYPE, NULL);
 #  endif
   if (locale == NULL || locale[0] == '\0')
@@ -248,7 +252,10 @@ locale_charset ()
 
 #endif
 
-  if (codeset != NULL && codeset[0] != '\0')
+  if (codeset == NULL)
+    /* The canonical name cannot be determined.  */
+    codeset = "";
+  else if (codeset[0] != '\0')
     {
       /* Resolve alias. */
       for (aliases = get_charset_aliases ();
diff --git a/lib/ChangeLog b/lib/ChangeLog
index 6dc88b5..97cc9ad 100644
--- a/lib/ChangeLog
+++ b/lib/ChangeLog
@@ -1,3 +1,8 @@
+2001-03-09  Bruno Haible  <haible@clisp.cons.org>
+
+	* config.charset: Update from libiconv-1.6.
+	* localcharset.c: Likewise.
+
 2001-03-04  Bruno Haible  <haible@clisp.cons.org>
 
 	* system.h (ISSLASH, HAS_DEVICE, IS_ABSOLUTE_PATH, IS_PATH_WITH_DIR,
diff --git a/lib/config.charset b/lib/config.charset
index bf21d99..0844094 100755
--- a/lib/config.charset
+++ b/lib/config.charset
@@ -33,27 +33,37 @@
 #       name                         used by which systems         a MIME name?
 #   ASCII, ANSI_X3.4-1968     glibc solaris freebsd
 #   ISO-8859-1                glibc aix hpux irix osf solaris freebsd   yes
-#   ISO-8859-2                glibc aix hpux irix solaris freebsd       yes
+#   ISO-8859-2                glibc aix hpux irix osf solaris freebsd   yes
 #   ISO-8859-3                glibc                                     yes
-#   ISO-8859-4                solaris freebsd                           yes
-#   ISO-8859-5                glibc aix hpux irix solaris freebsd       yes
+#   ISO-8859-4                osf solaris freebsd                       yes
+#   ISO-8859-5                glibc aix hpux irix osf solaris freebsd   yes
 #   ISO-8859-6                glibc aix hpux solaris                    yes
 #   ISO-8859-7                glibc aix hpux irix osf solaris           yes
-#   ISO-8859-8                glibc aix hpux solaris                    yes
+#   ISO-8859-8                glibc aix hpux osf solaris                yes
 #   ISO-8859-9                glibc aix hpux irix osf solaris           yes
 #   ISO-8859-13               glibc
-#   ISO-8859-15               glibc aix solaris freebsd
+#   ISO-8859-15               glibc aix osf solaris freebsd
 #   KOI8-R                    glibc solaris freebsd                     yes
 #   KOI8-U                    glibc freebsd                             yes
-#   CP850                     aix osf
+#   CP437                     dos
+#   CP775                     dos
+#   CP850                     aix osf dos
+#   CP852                     dos
+#   CP855                     dos
 #   CP856                     aix
-#   CP866                     freebsd
-#   CP874                     win32
+#   CP857                     dos
+#   CP861                     dos
+#   CP862                     dos
+#   CP864                     dos
+#   CP865                     dos
+#   CP866                     freebsd dos
+#   CP869                     dos
+#   CP874                     win32 dos
 #   CP922                     aix
-#   CP932                     aix win32
+#   CP932                     aix win32 dos
 #   CP943                     aix
-#   CP949                     win32
-#   CP950                     win32
+#   CP949                     osf win32 dos
+#   CP950                     win32 dos
 #   CP1046                    aix
 #   CP1124                    aix
 #   CP1129                    aix
@@ -66,16 +76,16 @@
 #   CP1256                    win32
 #   CP1257                    win32
 #   GB2312                    glibc aix hpux irix solaris freebsd       yes
-#   EUC-JP                    glibc aix hpux irix solaris freebsd       yes
-#   EUC-KR                    glibc aix hpux irix solaris freebsd       yes
-#   EUC-TW                    glibc aix hpux irix solaris
-#   BIG5                      glibc aix hpux solaris freebsd            yes
+#   EUC-JP                    glibc aix hpux irix osf solaris freebsd   yes
+#   EUC-KR                    glibc aix hpux irix osf solaris freebsd   yes
+#   EUC-TW                    glibc aix hpux irix osf solaris
+#   BIG5                      glibc aix hpux osf solaris freebsd        yes
 #   BIG5HKSCS                 glibc
-#   GBK                       aix win32
+#   GBK                       aix osf win32 dos
 #   GB18030                   glibc
-#   SJIS                      hpux solaris freebsd
+#   SJIS                      hpux osf solaris freebsd
 #   JOHAB                     glibc win32
-#   TIS-620                   glibc aix hpux solaris
+#   TIS-620                   glibc aix hpux osf solaris
 #   VISCII                    glibc                                     yes
 #   HP-ROMAN8                 hpux
 #   HP-ARABIC8                hpux
@@ -83,7 +93,9 @@
 #   HP-HEBREW8                hpux
 #   HP-TURKISH8               hpux
 #   HP-KANA8                  hpux
-#   UTF-8                     glibc aix hpux solaris                    yes
+#   DEC-KANJI                 osf
+#   DEC-HANYU                 osf
+#   UTF-8                     glibc aix hpux osf solaris                yes
 #
 # Note: Names which are not marked as being a MIME name should not be used in
 # Internet protocols for information interchange (mail, news, etc.).
@@ -178,9 +190,28 @@ case "$os" in
 	;;
     osf*)
 	echo "ISO8859-1 ISO-8859-1"
+	echo "ISO8859-2 ISO-8859-2"
+	echo "ISO8859-4 ISO-8859-4"
+	echo "ISO8859-5 ISO-8859-5"
 	echo "ISO8859-7 ISO-8859-7"
+	echo "ISO8859-8 ISO-8859-8"
 	echo "ISO8859-9 ISO-8859-9"
+	echo "ISO8859-15 ISO-8859-15"
 	echo "cp850 CP850"
+	echo "big5 BIG5"
+	echo "dechanyu DEC-HANYU"
+	echo "dechanzi GB2312"
+	echo "deckanji DEC-KANJI"
+	echo "deckorean EUC-KR"
+	echo "eucJP EUC-JP"
+	echo "eucKR EUC-KR"
+	echo "eucTW EUC-TW"
+	echo "GBK GBK"
+	echo "KSC5601 CP949"
+	echo "sdeckanji EUC-JP"
+	echo "SJIS SJIS"
+	echo "TACTIS TIS-620"
+	echo "UTF-8 UTF-8"
 	;;
     solaris*)
 	echo "646 ASCII"
@@ -214,8 +245,8 @@ case "$os" in
 	  echo "$l.ASCII ASCII"
 	done
 	for l in da_DK de_AT de_CH de_DE en_AU en_CA en_GB en_US es_ES \
-                 fi_FI fr_BE fr_CA fr_CH fr_FR is_IS it_CH it_IT la_LN \
-                 lt_LN nl_BE nl_NL no_NO pt_PT sv_SE; do
+	         fi_FI fr_BE fr_CA fr_CH fr_FR is_IS it_CH it_IT la_LN \
+	         lt_LN nl_BE nl_NL no_NO pt_PT sv_SE; do
 	  echo "$l.ISO_8859-1 ISO-8859-1"
 	  echo "$l.DIS_8859-15 ISO-8859-15"
 	done
@@ -243,4 +274,165 @@ case "$os" in
 	# BeOS has a single locale, and it has UTF-8 encoding.
 	echo "C UTF-8"
 	;;
+    msdosdjgpp*)
+	# DJGPP 2.03 doesn't have nl_langinfo(CODESET); therefore
+	# localcharset.c falls back to using the full locale name
+	# from the environment variables.
+	echo "#"
+	echo "# The encodings given here may not all be correct."
+	echo "# If you find that the encoding given for your language and"
+	echo "# country is not the one your DOS machine actually uses, just"
+	echo "# correct it in this file, and send a mail to"
+	echo "# Juan Manuel Guerrero <st001906@hrz1.hrz.tu-darmstadt.de>"
+	echo "# and Bruno Haible <haible@clisp.cons.org>."
+	echo "#"
+	echo "C ASCII"
+	# ISO-8859-1 languages
+	echo "ca CP850"
+	echo "ca_ES CP850"
+	echo "da CP865"    # not CP850 ??
+	echo "da_DK CP865" # not CP850 ??
+	echo "de CP850"
+	echo "de_AT CP850"
+	echo "de_CH CP850"
+	echo "de_DE CP850"
+	echo "en CP850"
+	echo "en_AU CP850" # not CP437 ??
+	echo "en_CA CP850"
+	echo "en_GB CP850"
+	echo "en_NZ CP437"
+	echo "en_US CP437"
+	echo "en_ZA CP850" # not CP437 ??
+	echo "es CP850"
+	echo "es_AR CP850"
+	echo "es_BO CP850"
+	echo "es_CL CP850"
+	echo "es_CO CP850"
+	echo "es_CR CP850"
+	echo "es_CU CP850"
+	echo "es_DO CP850"
+	echo "es_EC CP850"
+	echo "es_ES CP850"
+	echo "es_GT CP850"
+	echo "es_HN CP850"
+	echo "es_MX CP850"
+	echo "es_NI CP850"
+	echo "es_PA CP850"
+	echo "es_PY CP850"
+	echo "es_PE CP850"
+	echo "es_SV CP850"
+	echo "es_UY CP850"
+	echo "es_VE CP850"
+	echo "et CP850"
+	echo "et_EE CP850"
+	echo "eu CP850"
+	echo "eu_ES CP850"
+	echo "fi CP850"
+	echo "fi_FI CP850"
+	echo "fr CP850"
+	echo "fr_BE CP850"
+	echo "fr_CA CP850"
+	echo "fr_CH CP850"
+	echo "fr_FR CP850"
+	echo "ga CP850"
+	echo "ga_IE CP850"
+	echo "gd CP850"
+	echo "gd_GB CP850"
+	echo "gl CP850"
+	echo "gl_ES CP850"
+	echo "id CP850"    # not CP437 ??
+	echo "id_ID CP850" # not CP437 ??
+	echo "is CP861"    # not CP850 ??
+	echo "is_IS CP861" # not CP850 ??
+	echo "it CP850"
+	echo "it_CH CP850"
+	echo "it_IT CP850"
+	echo "lt CP775"
+	echo "lt_LT CP775"
+	echo "lv CP775"
+	echo "lv_LV CP775"
+	echo "nb CP865"    # not CP850 ??
+	echo "nb_NO CP865" # not CP850 ??
+	echo "nl CP850"
+	echo "nl_BE CP850"
+	echo "nl_NL CP850"
+	echo "nn CP865"    # not CP850 ??
+	echo "nn_NO CP865" # not CP850 ??
+	echo "no CP865"    # not CP850 ??
+	echo "no_NO CP865" # not CP850 ??
+	echo "pt CP850"
+	echo "pt_BR CP850"
+	echo "pt_PT CP850"
+	echo "sv CP850"
+	echo "sv_SE CP850"
+	# ISO-8859-2 languages
+	echo "cs CP852"
+	echo "cs_CZ CP852"
+	echo "hr CP852"
+	echo "hr_HR CP852"
+	echo "hu CP852"
+	echo "hu_HU CP852"
+	echo "pl CP852"
+	echo "pl_PL CP852"
+	echo "ro CP852"
+	echo "ro_RO CP852"
+	echo "sk CP852"
+	echo "sk_SK CP852"
+	echo "sl CP852"
+	echo "sl_SI CP852"
+	echo "sq CP852"
+	echo "sq_AL CP852"
+	echo "sr CP852"    # CP852 or CP866 or CP855 ??
+	echo "sr_YU CP852" # CP852 or CP866 or CP855 ??
+	# ISO-8859-3 languages
+	echo "mt CP850"
+	echo "mt_MT CP850"
+	# ISO-8859-5 languages
+	echo "be CP866"
+	echo "be_BE CP866"
+	echo "bg CP866"    # not CP855 ??
+	echo "bg_BG CP866" # not CP855 ??
+	echo "mk CP866"    # not CP855 ??
+	echo "mk_MK CP866" # not CP855 ??
+	echo "ru KOI8-R"    # not CP866 ??
+	echo "ru_RU KOI8-R" # not CP866 ??
+	# ISO-8859-6 languages
+	echo "ar CP864"
+	echo "ar_AE CP864"
+	echo "ar_DZ CP864"
+	echo "ar_EG CP864"
+	echo "ar_IQ CP864"
+	echo "ar_IR CP864"
+	echo "ar_JO CP864"
+	echo "ar_KW CP864"
+	echo "ar_MA CP864"
+	echo "ar_OM CP864"
+	echo "ar_QA CP864"
+	echo "ar_SA CP864"
+	echo "ar_SY CP864"
+	# ISO-8859-7 languages
+	echo "el CP869"
+	echo "el_GR CP869"
+	# ISO-8859-8 languages
+	echo "he CP862"
+	echo "he_IL CP862"
+	# ISO-8859-9 languages
+	echo "tr CP857"
+	echo "tr_TR CP857"
+	# Japanese
+	echo "ja CP932"
+	echo "ja_JP CP932"
+	# Chinese
+	echo "zh_CN GBK"
+	echo "zh_TW CP950" # not CP938 ??
+	# Korean
+	echo "kr CP949"    # not CP934 ??
+	echo "kr_KR CP949" # not CP934 ??
+	# Thai
+	echo "th CP874"
+	echo "th_TH CP874"
+	# Other
+	echo "eo CP850"
+	echo "eo_EO CP850"
+	;;
 esac
diff --git a/lib/localcharset.c b/lib/localcharset.c
index 12214b7..201483c 100644
--- a/lib/localcharset.c
+++ b/lib/localcharset.c
@@ -1,6 +1,6 @@
 /* Determine a canonical name for the current locale's character encoding.
 
-   Copyright (C) 2000, 2001 Free Software Foundation, Inc.
+   Copyright (C) 2000-2001 Free Software Foundation, Inc.
 
    This program is free software; you can redistribute it and/or modify it
    under the terms of the GNU Library General Public License as published
@@ -194,7 +194,7 @@ get_charset_aliases ()
    into one of the canonical names listed in config.charset.
    The result must not be freed; it is statically allocated.
    If the canonical name cannot be determined, the result is a non-canonical
-   name or NULL.  */
+   name.  */
 
 #ifdef STATIC
 STATIC
@@ -214,10 +214,14 @@ locale_charset ()
 
 # else
 
-  /* On old systems which lack it, use setlocale and getenv.  */
+  /* On old systems which lack it, use setlocale or getenv.  */
   const char *locale = NULL;
 
-#  if HAVE_SETLOCALE
+  /* But most old systems don't have a complete set of locales.  Some
+     (like SunOS 4 or DJGPP) have only the C locale.  Therefore we don't
+     use setlocale here; it would return "C" when it doesn't support the
+     locale name the user has set.  */
+#  if HAVE_SETLOCALE && 0
   locale = setlocale (LC_CTYPE, NULL);
 #  endif
   if (locale == NULL || locale[0] == '\0')
@@ -248,7 +252,10 @@ locale_charset ()
 
 #endif
 
-  if (codeset != NULL && codeset[0] != '\0')
+  if (codeset == NULL)
+    /* The canonical name cannot be determined.  */
+    codeset = "";
+  else if (codeset[0] != '\0')
     {
       /* Resolve alias. */
       for (aliases = get_charset_aliases ();