Document the problem with the swash_fetch() API that affects

more complex case conversions.

p4raw-id: //depot/perl@12450

diff --git a/lib/utf8_heavy.pl b/lib/utf8_heavy.pl
index bbc082b..fe286d9 100644 (file)
--- a/lib/utf8_heavy.pl
+++ b/lib/utf8_heavy.pl
@@ -159,6 +159,7 @@ sub SWASHNEW {
 # NOTE: utf8.c:swash_init() assumes entries are never modified once generated.
 
 sub SWASHGET {
+    # See utf8.c:Perl_swash_fetch for problems with this interface.
     my ($self, $start, $len) = @_;
     local $^D = 0 if $^D;
     my $type = $self->{TYPE};

diff --git a/utf8.c b/utf8.c
index e1a7e63..0c09469 100644 (file)
--- a/utf8.c
+++ b/utf8.c
@@ -1281,6 +1281,12 @@ Perl_swash_init(pTHX_ char* pkg, char* name, SV *listsv, I32 minbits, I32 none)
     return retval;
 }
 
+
+/* This API is wrong for special case conversions since we may need to
+ * return several Unicode characters for a single Unicode character
+ * (see lib/unicore/SpecCase.txt) The SWASHGET in lib/utf8_heavy.pl is
+ * the lower-level routine, and it is similarly broken for returning
+ * multiple values.  --jhi */
 UV
 Perl_swash_fetch(pTHX_ SV *sv, U8 *ptr, bool do_utf8)
 {