Re: Boost.Compute broken in 1.86.0 because of UUID changes, 1.86.1?

From: Peter Dimov (pdimov_at_[hidden])
Date: 2024-08-22 09:54:05

Boris Kolpackov wrote:
> Peter Dimov via Boost <boost_at_[hidden]> writes:
> 
> > Because Boost.Compute (boost/compute/detail/sha1.hpp) uses an
> > implementation detail of Boost.UUID (boost/uuid/detail/sha1.hpp), and
> > because I changed the implementation to fix an endianness issue,
> > <boost/compute.hpp> now causes a compile error in 1.86.0
> 
> Any sense why this was not detected (e.g., by CI) when the change was made
> or during pre-release testing? It seems to me compiling any translation unit
> that included <boost/compute.hpp> would have picked this up, no?

It would have, but no such translation unit was compiled by anyone before
the release went out.

There's no CI that runs the entire Boost test suite (that would take many hours)
but even if there were, Boost.Compute's tests require OpenCL and automatically
disable themselves when it's not present.

This is why, in principle, we do a beta release, to catch these issues before the
real release, but it doesn't seem to be effective.