Nathan Beyer said the following on 02.04.2008 5:47:
On Tue, Apr 1, 2008 at 8:21 PM, Nathan Beyer <[EMAIL PROTECTED]> wrote:
On Tue, Apr 1, 2008 at 1:26 PM, Gregory Shimansky <[EMAIL PROTECTED]> wrote:
 > On 1 April 2008 Nathan Beyer wrote:
 >  > I've just finished testing on Win x86 with the new apr code and no
 >  > patches. All of the kernel.test, smoke.test, reg.test and cunit.test
 >  > pass without failure or error.
 >  >
 >  > If there's no objection, I'm going to update to the latest apr source
 >  > and remove the patches tonight.
 >
 >  Just to make sure, did you define some environment variable with no string
 >  value ("" empty string) before running kernel tests?
 >
 >
 No - i'll give that a try and see what happens.
 -Nathan

I must be an idiot, but how do you do this? On WinXP, when I "set
BAD_VAR=", nothing is added. When I set an existing variable, it's
removed from the environment. Is this something that can be done on
other versions of Windows?

Hmm indeed. The funny thing I is that I cannot reproduce it myself. Looks like such variables are not allowed to be set in legal ways. But I do remember that windows set that PRINTER variable to have no value when no printer was defined...

Ok, I think we are safe to go with APR 1.2.12. If someone encounters this bug again we'll know where to look.



 >  > On Mon, Mar 31, 2008 at 9:02 PM, Nathan Beyer <[EMAIL PROTECTED]> wrote:
 >  > > I've tested apr 1.2.12, apr-util 1.2.12, apr-iconv 1.2.1 on Linux
 >  > >  x86_64 with the patched unix code removed and the DRLVM tests seem to
 >  > >  all be passing. I'm running the tests again to be sure. I'm also
 >  > >  running the tests on Linux x86.
 >  > >
 >  > >  On Thu, Mar 27, 2008 at 9:22 PM, Nathan Beyer <[EMAIL PROTECTED]> 
wrote:
 >  > >  > Is there any specific reason we're still on apr/apr-util 1.2.6?
 >  > >  >
 >  > >  >  I'm trying to get things running on Leopard and it seems 1.2.12 is
 >  > >  >  necessity. How do we want to approach such an upgrade?
 >  > >  >
 >  > >  >  I noticed that the apr-iconv is also a bit out of date.
 >  > >  >
 >  > >  >  -Nathan
 >
 >  --
 >  Gregory
 >



--
Gregory

Reply via email to