From: Matt L. <mat...@gm...> - 2011-02-25 16:29:36
|
On Fri, Feb 25, 2011 at 11:06 AM, Derek Parnell <ddp...@bi...> wrote: > On Sat, 26 Feb 2011 00:01:24 +1100, Matt Lewis > <mat...@gm...> wrote: > >> There are still a few open bugs, and a few marked as possibly fixed >> that need confirmation. Several of these are windows specific. Can >> some folks on windows take a look and close them out? > > I'm getting a lot of redefinition errors from watcom of pcre identifiers. I get similar issues sometimes when updating and not cleaning. Ideally, dependencies would take care of that, but it doesn't always seem to happen. > >> There are two issues that I think we really need to get figured out: >> >> 600: Debugger color clash > > This is not a problem in Windows. Yes, I have a feeling it may be another go around with the different terminals using different colors. > It seems that conversion of floating point numbers has been changed. > > This used to work fine ... > > constant x = 2.8844991406148167646432766215602 > > but now I have to shorten it to ... > > constant x = 2.8844991406148167646432 > > > Is this a bug? Yes, I would think so. The extra precision should simply be ignored. Also, the 64-bit version will probably eventually support more precision, so we'd need to support it in that case, too. > Also, it used to be that if a unittest case failed, eutest would just > report it and go on to the next test. Now it seems that when a test fails, > eutest skips the rest of the tests in the file. Is this actually the case > now? If so, I feel that it is a mistake because it slows down the testing > process. I haven't noticed this. -- Matt Lewis |