On 7/9/2014 11:21 AM, Beman Dawes wrote:
On Wed, Jul 9, 2014 at 2:24 AM, Eric Niebler wrote:
On 07/08/2014 02:07 PM, Beman Dawes wrote:
Marshall is just a few days away from being able to put out a beta release candidate. (And that is very good news!)
But we still have too many libraries failing across the board on master:
accumulators
Two tests broke long ago due to some change in Boost.Random, which is used only by the tests. I've never found time to fix it. I could take these broken tests out of the matrix, if it helps.
Or markup the expected failures master status/explicit-failures-markup.xml so the reports tell release managers and users what is going on.
Done.
<sigh>Someday we will have an automatic testing system that rejects changes which break other libraries.</sigh>
I think that would be too restrictive. Sometimes we break interfaces. This happens to be one of those cases.
proto
Thanks for the heads up. I tend not to look at the tests unless I change something, but stuff tends to break all by itself. I'll add it to the work queue.
This failure was caused by the following change to boost::ref by Agustín Bergé (K-ballo): https://github.com/boostorg/core/commit/af629ffa59094048c335609f285afe342fd1... I didn't see any discussion about this change. I can guess why it was made: to make boost::ref behave like std::ref, which does "reference collapsing" with reference_wrapper (which IMO is broken, and I unsuccessfully argued that in the committee when this was voted in). The problem is, it's a breaking interface change to one of the oldest, most heavily used, and stable pieces of Boost. It pretty massively breaks a major part of Boost.Proto's interface (proto::make_expr). I've hacked around the problem in Proto's tests, but this is going to break end-user code, and I don't know what I'm going to tell people. I don't like differences between boost and std. But I also don't like breaking code. If it were up to me, I'd probably back out the change to boost::ref for 1.56 and have a discussion. <sigh> \e