Update.

	* locales/zh_CN: Remove non-ASCII digit entry.
	* locales/ja_JP: Likewise.

2000-09-30  Bruno Haible  <haible@clisp.cons.org>

	* gen-unicode-ctype.c (is_digit, is_xdigit): Remove all non-ASCII
	digits.
	(is_alpha): Add them here.
	* locales/i18n (digit): Remove all non-ASCII digits.
	(alpha): Add them here.

	* dump-ctype.c: Pretty print.

2000-10-04  Ulrich Drepper  <drepper@redhat.com>

diff --git a/localedata/ChangeLog b/localedata/ChangeLog
index 30cbff3..296a5ba 100644 (file)
--- a/localedata/ChangeLog
+++ b/localedata/ChangeLog
@@ -1,5 +1,20 @@
 2000-10-04  Ulrich Drepper  <drepper@redhat.com>
 
+       * locales/zh_CN: Remove non-ASCII digit entry.
+       * locales/ja_JP: Likewise.
+
+2000-09-30  Bruno Haible  <haible@clisp.cons.org>
+
+       * gen-unicode-ctype.c (is_digit, is_xdigit): Remove all non-ASCII
+       digits.
+       (is_alpha): Add them here.
+       * locales/i18n (digit): Remove all non-ASCII digits.
+       (alpha): Add them here.
+
+       * dump-ctype.c: Pretty print.
+
+2000-10-04  Ulrich Drepper  <drepper@redhat.com>
+
        * charmaps/EUC-KR: Change \x5c mapping to U20a9.
        * locales/ko_KR: Change currency_symbol back to use U20a9.
 

diff --git a/localedata/dump-ctype.c b/localedata/dump-ctype.c
index 9514ca7..66d539a 100644 (file)
--- a/localedata/dump-ctype.c
+++ b/localedata/dump-ctype.c
@@ -113,7 +113,8 @@ static void dump_map (const char *map_name)
     }
 }
 
-int main (int argc, char *argv[])
+int
+main (int argc, char *argv[])
 {
   size_t i;
 

diff --git a/localedata/gen-unicode-ctype.c b/localedata/gen-unicode-ctype.c
index b45e703..1259aef 100644 (file)
--- a/localedata/gen-unicode-ctype.c
+++ b/localedata/gen-unicode-ctype.c
@@ -166,21 +166,21 @@ fill_attributes (const char *unicodedata_filename)
       int n;
 
       lineno++;
-      n = getfield(stream, field0, ';');
-      n += getfield(stream, field1, ';');
-      n += getfield(stream, field2, ';');
-      n += getfield(stream, field3, ';');
-      n += getfield(stream, field4, ';');
-      n += getfield(stream, field5, ';');
-      n += getfield(stream, field6, ';');
-      n += getfield(stream, field7, ';');
-      n += getfield(stream, field8, ';');
-      n += getfield(stream, field9, ';');
-      n += getfield(stream, field10, ';');
-      n += getfield(stream, field11, ';');
-      n += getfield(stream, field12, ';');
-      n += getfield(stream, field13, ';');
-      n += getfield(stream, field14, '\n');
+      n = getfield (stream, field0, ';');
+      n += getfield (stream, field1, ';');
+      n += getfield (stream, field2, ';');
+      n += getfield (stream, field3, ';');
+      n += getfield (stream, field4, ';');
+      n += getfield (stream, field5, ';');
+      n += getfield (stream, field6, ';');
+      n += getfield (stream, field7, ';');
+      n += getfield (stream, field8, ';');
+      n += getfield (stream, field9, ';');
+      n += getfield (stream, field10, ';');
+      n += getfield (stream, field11, ';');
+      n += getfield (stream, field12, ';');
+      n += getfield (stream, field13, ';');
+      n += getfield (stream, field14, '\n');
       if (n == 0)
        break;
       if (n != 15)
@@ -196,21 +196,21 @@ fill_attributes (const char *unicodedata_filename)
        {
          /* Deal with a range. */
          lineno++;
-         n = getfield(stream, field0, ';');
-         n += getfield(stream, field1, ';');
-         n += getfield(stream, field2, ';');
-         n += getfield(stream, field3, ';');
-         n += getfield(stream, field4, ';');
-         n += getfield(stream, field5, ';');
-         n += getfield(stream, field6, ';');
-         n += getfield(stream, field7, ';');
-         n += getfield(stream, field8, ';');
-         n += getfield(stream, field9, ';');
-         n += getfield(stream, field10, ';');
-         n += getfield(stream, field11, ';');
-         n += getfield(stream, field12, ';');
-         n += getfield(stream, field13, ';');
-         n += getfield(stream, field14, '\n');
+         n = getfield (stream, field0, ';');
+         n += getfield (stream, field1, ';');
+         n += getfield (stream, field2, ';');
+         n += getfield (stream, field3, ';');
+         n += getfield (stream, field4, ';');
+         n += getfield (stream, field5, ';');
+         n += getfield (stream, field6, ';');
+         n += getfield (stream, field7, ';');
+         n += getfield (stream, field8, ';');
+         n += getfield (stream, field9, ';');
+         n += getfield (stream, field10, ';');
+         n += getfield (stream, field11, ';');
+         n += getfield (stream, field12, ';');
+         n += getfield (stream, field13, ';');
+         n += getfield (stream, field14, '\n');
          if (n != 15)
            {
              fprintf (stderr, "missing end range in '%s':%d\n",
@@ -390,17 +390,35 @@ is_alpha (unsigned int ch)
              || (unicode_attributes[ch].category[0] == 'S'
                  && unicode_attributes[ch].category[1] == 'o'
                  && strstr (unicode_attributes[ch].name, " LETTER ")
-                    != NULL)));
+                    != NULL)
+             /* Consider all the non-ASCII digits as alphabetic.
+                ISO C 99 forbids us to have them in category "digit",
+                but we want iswalnum to return true on them.  */
+             || (unicode_attributes[ch].category[0] == 'N'
+                 && unicode_attributes[ch].category[1] == 'd'
+                 && !(ch >= 0x0030 && ch <= 0x0039))));
 }
 
 static bool
 is_digit (unsigned int ch)
 {
+#if 0
   return (unicode_attributes[ch].name != NULL
          && unicode_attributes[ch].category[0] == 'N'
          && unicode_attributes[ch].category[1] == 'd');
   /* Note: U+0BE7..U+0BEF and U+1369..U+1371 are digit systems without
      a zero.  Must add <0> in front of them by hand.  */
+#else
+  /* SUSV2 gives us some freedom for the "digit" category, but ISO C 99
+     takes it away:
+     7.25.2.1.5:
+        The iswdigit function tests for any wide character that corresponds
+        to a decimal-digit character (as defined in 5.2.1).
+     5.2.1:
+        the 10 decimal digits 0 1 2 3 4 5 6 7 8 9
+   */
+  return (ch >= 0x0030 && ch <= 0x0039);
+#endif
 }
 
 static bool
@@ -455,9 +473,23 @@ is_cntrl (unsigned int ch)
 static bool
 is_xdigit (unsigned int ch)
 {
+#if 0
   return is_digit (ch)
         || (ch >= 0x0041 && ch <= 0x0046)
         || (ch >= 0x0061 && ch <= 0x0066);
+#else
+  /* SUSV2 gives us some freedom for the "xdigit" category, but ISO C 99
+     takes it away:
+     7.25.2.1.12:
+        The iswxdigit function tests for any wide character that corresponds
+        to a hexadecimal-digit character (as defined in 6.4.4.1).
+     6.4.4.1:
+        hexadecimal-digit: one of 0 1 2 3 4 5 6 7 8 9 a b c d e f A B C D E F
+   */
+  return (ch >= 0x0030 && ch <= 0x0039)
+        || (ch >= 0x0041 && ch <= 0x0046)
+        || (ch >= 0x0061 && ch <= 0x0066);
+#endif
 }
 
 static bool

diff --git a/localedata/locales/i18n b/localedata/locales/i18n
index b284ded..977c585 100644 (file)
--- a/localedata/locales/i18n
+++ b/localedata/locales/i18n
@@ -305,13 +305,10 @@ alpha /
    <UFE70>..<UFE72>;<UFE74>;<UFE76>..<UFEFC>;/
 % HALFWIDTH AND FULLWIDTH FORMS/
    <UFF21>..<UFF3A>;<UFF41>..<UFF5A>;<UFF66>..<UFFBE>;<UFFC2>..<UFFC7>;/
-   <UFFCA>..<UFFCF>;<UFFD2>..<UFFD7>;<UFFDA>..<UFFDC>
-
-% The "digit" class of the "i18n" FDCC-set is reflecting
-% the recommendations in TR 10176 annex A
-digit /
-% TABLE 1 BASIC LATIN/
-  <U0030>..<U0039>;/
+   <UFFCA>..<UFFCF>;<UFFD2>..<UFFD7>;<UFFDA>..<UFFDC>;/
+% The non-ASCII number characters are included here because ISO C 99    /
+% forbids us to classify them as digits; however, they behave more like /
+% alphanumeric than like punctuation.                                   /
 % TABLE 15 and 16 ARABIC/
   <U0660>..<U0669>;<U06F0>..<U06F9>;/
 % TABLE 17 DEVANAGARI/
@@ -349,6 +346,11 @@ digit /
 % HALFWIDTH AND FULLWIDTH FORMS/
    <UFF10>..<UFF19>
 
+% The "digit" class must only contain the BASIC LATIN digits, says ISO C 99
+% (sections 7.25.2.1.5 and 5.2.1).
+digit /
+   <U0030>..<U0039>
+
 outdigit <U0030>..<U0039>
 
 space /
@@ -602,6 +604,8 @@ print /
    <UFFD2>..<UFFD7>;<UFFDA>..<UFFDC>;<UFFE0>..<UFFE6>;<UFFE8>..<UFFEE>;/
    <UFFF9>..<UFFFD>
 
+% The "xdigit" class must only contain the BASIC LATIN digits and A-F, a-f,
+% says ISO C 99 (sections 7.25.2.1.12 and 6.4.4.1).
 xdigit /
    <U0030>..<U0039>;<U0041>..<U0046>;<U0061>..<U0066>
 

diff --git a/localedata/locales/ja_JP b/localedata/locales/ja_JP
index 4c4d2b6..826247f 100644 (file)
--- a/localedata/locales/ja_JP
+++ b/localedata/locales/ja_JP
@@ -155,9 +155,7 @@ alpha       <U0041>;<U0042>;<U0043>;<U0044>;<U0045>;<U0046>;<U0047>;<U0048>;/
        <U0175>;<U00FD>;<U00FF>;<U0177>;<U017A>;<U017E>;<U017C>
 
 digit  <U0030>;<U0031>;<U0032>;<U0033>;<U0034>;/
-       <U0035>;<U0036>;<U0037>;<U0038>;<U0039>;/
-       <UFF10>;<UFF11>;<UFF12>;<UFF13>;<UFF14>;/
-       <UFF15>;<UFF16>;<UFF17>;<UFF18>;<UFF19>
+       <U0035>;<U0036>;<U0037>;<U0038>;<U0039>
 
 xdigit <U0030>;<U0031>;<U0032>;<U0033>;<U0034>;/
        <U0035>;<U0036>;<U0037>;<U0038>;<U0039>;/

diff --git a/localedata/locales/zh_CN b/localedata/locales/zh_CN
index 7c98bc9..278dd23 100644 (file)
--- a/localedata/locales/zh_CN
+++ b/localedata/locales/zh_CN
@@ -90,9 +90,7 @@ alpha         <U0041>..<U005A>;<U0061>..<U007A>;/
 % Bopomofo/
        <U3105>..<U3129>
 
-digit   <U0030>..<U0039>;/
-% Fullwidth digits/
-       <UFF10>..<UFF19>
+digit   <U0030>..<U0039>
 
 outdigit <U0030>..<U0039>