On Wed, Jul 11, 2001 at 01:16:30AM -0700, Vadim Konovalov wrote: > > The patch I supplied and Jarkko committed was in toke.c, not just for > > QNX (was it?) so should solve Borland problems for bleadperl. > > It does not solved the problem, probably because problem is deeper in > Borland case: > > (for perl@11148) > Failed Test Stat Wstat Total Fail Failed List of Failed > ---------------------------------------------------------------------------- > --- > op/arith.t 109 4 3.67% 17-18 24 54 > (3 subtests UNEXPECTEDLY SUCCEEDED), 38 tests and 177 subtests skipped. > Failed 1/422 test scripts, 99.76% okay. 4/23037 subtests failed, 99.98% > okay. > > > > So I'm a bit confused - what still needs doing? > > Ok, many things are clearier now, I'll trace that problem again, will tell > you what is currently wrong and will try to fix it. I think the change is this one: [ 11190] By: jhi on 2001/07/07 15:22:08 Log: Subject: [PATCH] Re: op/arith.t From: Nicholas Clark <nick@ccl4.org> Date: Sat, 7 Jul 2001 13:04:54 +0100 Message-ID: <20010707130454.D59620@plum.flirble.org> Branch: perl ! numeric.c toke.c which is later than the 11148 you tested on. Nicholas ClarkThread Previous | Thread Next