On Tue, 8 Oct 2002, Brent Dax wrote:
> OK. This is a pretty big patch. (Fine--*really* big, at 117K. :^) )
>
> First of all, it completes the feature set of the Parrot_sprintf family,
Thanks!
After just this little patch: (line numbers are off due to other
unrelated fiddling on my part)
--- parrot-orig/misc.c Tue Oct 8 12:21:12 2002
+++ parrot-andy/misc.c Tue Oct 8 12:19:53 2002
@@ -612,7 +622,7 @@
break;
case 'p':
- chptr = va_arg(*args, void *);
+ chptr = va_arg(args, void *);
int_to_str(t1, t2, (HUGEINTVAL)chptr, 16);
handle_flags(&info, t1, 1, "0x");
it compiles. So far, I only have seen 2 failed tests:
The first is t/src/sprintf.t, which looks like an off-by-one issue of some
sort. I haven't tracked it further yet.
not ok 1 - hello world
# Failed test (t/src/sprintf.t at line 9)
# got: 'Hello, Parrot!
# PerlHash[0x100]
# PerlHash[0x100]
# Hello, Par!
# '
# expected: 'Hello, Parrot!
# PerlHash[0x100]
# PerlHash[0x100]
# Hello, Pa!
# '
# Looks like you failed 1 tests of 1.
The second is t/op/string.t, test 89:
not ok 89 - num to string
# Failed test (t/op/string.t at line 1260)
# got: ''
# expected: '80.43
# -1.11111
# '
There were also some compiler warnings that I'll try to track down
further, and I haven't tested things exhaustively, but this sure looks
like the way to go to me. Thanks.
--
Andy Dougherty [EMAIL PROTECTED]
