|
Boost Testing : |
From: David Abrahams (dave_at_[hidden])
Date: 2007-08-21 13:52:57
on Tue Aug 21 2007, Christopher Cambly <ccambly-AT-ca.ibm.com> wrote:
> doug.gregor_at_[hidden] wrote on 08/21/2007 11:52:43 AM:
>
>> The Subversion tarball over at Meta-comm hasn't been updated in a very
>> long time, so all of our regression tests are out-of-date (and some,
>> like I just reported to the main developer's list, are hanging
>> machines).
>>
>> So, my first question: why are we relying on tarballs at all to get
>> Boost from Subversion? It seems like we would be much better over just
>> having the script do an "svn checkout" once (via "http", of course,
>> not "https") and then "svn up" before running the tests. That should
>> eliminate one of our failure points...
>>
>> - Doug
>
>>From our side, we prefer the tarball since we cannot install SVN on our
> machines without a legal review and approval of SVN itself. We are an
> oddity on this point, but it truly saves us a ton of hassle having
> a tarball available.
http://boost-consulting.com/boost.tar.bz2
-- Dave Abrahams Boost Consulting http://www.boost-consulting.com The Astoria Seminar ==> http://www.astoriaseminar.com