Change 16550 by jhi@alpha on 2002/05/11 22:24:31
The toke.c wariness about UTF-8 (really, UTF-EBCDIC) source
code in EBCDIC does not apply any more. (The utf8.t needed
a little bit of relaxing to work in EBCDIC, since *both* the
bytes in subtest #30 are illegal UTF-EBCDIC, and the error
message matched is slight different.) The utf8.t now passes.
Affected files ...
.... //depot/perl/lib/utf8.t#8 edit
.... //depot/perl/toke.c#432 edit
Differences ...
==== //depot/perl/lib/utf8.t#8 (text) ====
Index: perl/lib/utf8.t
--- perl/lib/utf8.t#7~15714~ Wed Apr 3 11:15:33 2002
+++ perl/lib/utf8.t Sat May 11 15:24:31 2002
@@ -166,7 +166,7 @@
use utf8; %a = ("\xE1\xA0"=>"sterling");
print 'start'; printf '%x,', ord \$_ foreach keys %a; print "end\n";
BANG
- qr/^Malformed UTF-8 character \(2 bytes, need 3.+\).*start\d+,end$/s
+ qr/^Malformed UTF-8 character \(\d bytes?, need \d,
+.+\).*start\d+,end$/sm
],
);
foreach (@tests) {
==== //depot/perl/toke.c#432 (text) ====
Index: perl/toke.c
--- perl/toke.c#431~16302~ Wed May 1 05:54:24 2002
+++ perl/toke.c Sat May 11 15:24:31 2002
@@ -41,11 +41,7 @@
#ifdef USE_UTF8_SCRIPTS
# define UTF (!IN_BYTES)
#else
-# ifdef EBCDIC /* For now 'use utf8' does not affect tokenizer on EBCDIC */
-# define UTF (PL_linestr && DO_UTF8(PL_linestr))
-# else
-# define UTF ((PL_linestr && DO_UTF8(PL_linestr)) || (PL_hints & HINT_UTF8))
-# endif
+# define UTF ((PL_linestr && DO_UTF8(PL_linestr)) || (PL_hints & HINT_UTF8))
#endif
/* In variables named $^X, these are the legal values for X.
End of Patch.
