Unicode:UCD Clarify pod for num()

charclass_invlists.inc

@@ -456662,7 +456662,7 @@ static const U8 WB_dfa_table[] = {
 #endif	/* defined(PERL_IN_REGEXEC_C) */
 
 /* Generated from:
- * 7229a97216f54f7d47d5cff56fc8dbc185dcfe40db20533f8034a1215af787fe lib/Unicode/UCD.pm
+ * b7f46fc1010fd83f5a678b268a23fef0142a18d0ab2a142edd0bb03328e667c3 lib/Unicode/UCD.pm
  * 764f420cedfc8b43d9fec251c957a5d55fc45d40f6573f162990ed1dce7e36e0 lib/unicore/ArabicShaping.txt
  * b8f32554c6f658821fb0ee742d21c5b1f2086b9bf13071fed04894b022f93d67 lib/unicore/BidiBrackets.txt
  * d7afdadd1bbd66f5a663ac0e8f7958f18fd9491fc0bc59ec5877cb82db71db7d lib/unicore/BidiMirroring.txt

lib/Unicode/UCD.pm

@@ -5,7 +5,7 @@ use warnings;
 no warnings 'surrogate';    # surrogates can be inputs to this
 use charnames ();
 
-our $VERSION = '0.80';
+our $VERSION = '0.81';
 
 sub DEBUG () { 0 }
 $|=1 if DEBUG;
@@ -2496,6 +2496,30 @@ match them.  A single-character string containing one of these digits will
 have its decimal value returned by C<num>, but any longer string containing
 only these digits will return C<undef>.
 
+To illustrate further, the Rumi numeric symbols were used in centuries past in
+and around North Africa and the Iberian peninsula.  In order to be able to
+digitize the many historical documents that use them, Unicode has encoded the
+set.  There is no character representing zero.  There are characters for one
+through nine, ten, twenty, and so forth. C<num> correctly returns the values
+of these in isolation.
+
+ my $rumi_one = num("\N{RUMI DIGIT ONE}");
+ my $rumi_two = num("\N{RUMI DIGIT TWO}");
+ my $rumi_twenty = num("\N{RUMI NUMBER TWENTY}");
+ say "$rumi_one $rumi_two $rumi_twenty";     # 1 2 20
+
+Because these do not follow modern decimal positional notation, stringing more
+than one of these together doesn't mean what you likely would think it means.
+So, C<num> correctly returns C<undef> if you try.   If you request the length
+of the valid initial substring in this case, that length would be one.
+
+ my $len;
+ my $value = num("\N{RUMI DIGIT ONE}\N{RUMI DIGIT TWO}", \$len);
+ say $len, " ", (defined $value) ? $value : "undef";  # 1 undef
+
+How to represent numbers like twelve gets complicated, and Unicode doesn't
+give any guidance, so C<num> can't either.
+
 Strings of multiple sub- and superscripts are not recognized as numbers.  You
 can use either of the compatibility decompositions in Unicode::Normalize to
 change these into digits, and then call C<num> on the result.

regexp_constants.h

@@ -29,7 +29,7 @@
 #define MAX_FOLD_FROMS 3
 
 /* Generated from:
- * 7229a97216f54f7d47d5cff56fc8dbc185dcfe40db20533f8034a1215af787fe lib/Unicode/UCD.pm
+ * b7f46fc1010fd83f5a678b268a23fef0142a18d0ab2a142edd0bb03328e667c3 lib/Unicode/UCD.pm
  * 764f420cedfc8b43d9fec251c957a5d55fc45d40f6573f162990ed1dce7e36e0 lib/unicore/ArabicShaping.txt
  * b8f32554c6f658821fb0ee742d21c5b1f2086b9bf13071fed04894b022f93d67 lib/unicore/BidiBrackets.txt
  * d7afdadd1bbd66f5a663ac0e8f7958f18fd9491fc0bc59ec5877cb82db71db7d lib/unicore/BidiMirroring.txt