Re: Master Branch is now open for bug fixes

From: Marshall Clow (mclow.lists_at_[hidden])
Date: 2023-08-02 13:45:55

On Aug 1, 2023, at 10:48 AM, Alain O' Miniussi <alain.miniussi_at_oca.eu> wrote:
> ----- On Aug 1, 2023, at 6:49 PM, Marshall Clow <mclow.lists_at_[hidden]> wrote:
>> On Aug 1, 2023, at 6:27 AM, Alain O' Miniussi <alain.miniussi_at_oca.eu> wrote:
>> Hi,
>> 
>> The following commits:
>> https://github.com/boostorg/mpi/compare/master...develop
>> 
>> - fixes documentation typos.
>> - add missing declaration in header files.
>> - add test code for ctest.
>> - reports in Boost.MPI somes changes made in Boost.serialization
>> - fixes some bugs.
>> 
>> The only new feature is that it is now possible to get the MPI version string
>> 
>> I would not qualify any of those changes as critical, but some people have been waiting more than a year for them because of me.
>> 
>> Can I merge them ?
> Some of these changes look pretty substantial - https://github.com/boostorg/mpi/commit/da602162b4b00e8e3134a991d0c836f10cfe0b9e for example, changes 54 files.
> 
> Are these the kind of changes we should be making on the day before a release?
> On the other hand, if I just do "git diff master..develop --name-status” I see far fewer changes.
> 
> So I guess I’m confused.
> 
> Hi,
> 
> Right, somehow I overlooked that one. That specific commit has been made by Robert Ramey, I guess to comply with changes made in Serialization.
> But it seems that, in serialization, those changes are still in develop. On the over hand, merged Boost.MPI did pass the test on my platform with the master branch of serialization...
> Maybe we should at least wait for Robert input and probably skip this one.

I’ve compared the master and develop branches manually.
If you think that these changes won’t cause any problems, then you can go ahead.

— Marshall