Le Wed, Apr 23, 2025 at 04:11:49PM +0200, Theo Buehler a écrit :
> On Sat, Apr 19, 2025 at 06:02:00PM -0400, Brad Smith wrote:
> > On Thu, Apr 17, 2025 at 03:34:41AM -0400, Brad Smith wrote:
> > > Here is an update to Boost 1.88.
> > > 
> > > 
> > > https://www.boost.org/users/history/version_1_88_0.html
> > > 
> > > 
> > > Could use a bulk build to see what things are like in the tree.
> >  
> > Here is a rev 2 diff. I didn't mean to include building with ports Clang.
> 
> I think we should update to boost 1.87 first, using the diff below.
> 
> A bulk on current by someone else wouldn't hurt - I have yet again been
> bitten by a clang update not happening and going back is too much pain.
> My bulk results are therefore not fully reliable.
> 
> The 1.87 diff knocks out:
> 
> cad/openscad          filesystem              -> update to git snapshot
> games/vcmi            io_service/others       -> update to 1.6.3
> math/rstudio          io_service              -> mark broken
> net/fastnetmon                io_service, resolver    -> remove
> net/icinga            (?) coroutines          -> mark broken
> net/kea,postgresql    signbit()               -> mark broken
> 
> In addition to updating/marking broken, we need these pending fixes
> (on ports):
> 
> cad/librecad
> cad/prusaslicer
> graphics/orthanc/server       (postgres and webviewer plugins need more love)
> graphics/simgear      (for boost 1.88)
> net/kea                       (for all flavors except kea,postgresql)
> x11/cegui
> 
> So here's the suggested diff for boost 1.87. It's your and landry's
> latest diff with some rebasing and a backport for luceneplusplus:
> 
> https://github.com/boostorg/smart_ptr/commit/e7433ba54596da97cb7859455cd37ca140305a9c
> 
> This is about all I am going do for the latest round of pain from this
> piece of software I don't use and will continue to stay away from.

fwiw, i fully support the update to 1.87, against which most of the work
has been done. sending the diff for 1.88 without caring about all the
work that already went in for 1.87 was a bit rude...

i'll mark rstudio broken, and remove fastnetmon. theo, you have my ok if
you want to commit your diff. i'll have a quick look at the pending
diffs sent to ports@.

Landry

Reply via email to