[Backport support/2.14] Bump Boost shipped for Windows to v1.87#10651
[Backport support/2.14] Bump Boost shipped for Windows to v1.87#10651yhabteab merged 9 commits intosupport/2.14from
Conversation
(cherry picked from commit 11ab869)
It was removed in Boost 1.87. (cherry picked from commit 7bd35d8)
(cherry picked from commit 011c679)
not just boost::coroutines::detail::forced_unwind. This is needed because as of Boost 1.87, boost::asio::spawn() uses Fiber, not Coroutine v1. boostorg/asio@df973a85ed69f021 This is safe because every actual exception shall inherit from std::exception. Except forced_unwind and its Fiber equivalent, so that `catch(const std::exception&)` doesn't catch them and only them. (cherry picked from commit 0662f2b)
(cherry picked from commit fb2b2e2)
Creating the string_view from the std::string (as returned by GetData()) uses the stored length instead of having to detect it by finding '\0'. (cherry picked from commit ccfc722)
Simply giving two entire call expressions for either Boost version greatly improves readability in my opinion. (cherry picked from commit d1d399f)
f isn't used otherwise in the function, so if possible, it can just be moved into the lambda, avoiding a copy. Co-authored-by: Alexander Aleksandrovič Klimov <alexander.klimov@icinga.com> (cherry picked from commit d3fae44)
yhabteab
changed the title
julianbrost
left a comment
julianbrost
left a comment
There was a problem hiding this comment.
Please add the missing mailmap entry so that the remaining CI check is happy as well. (Apart from that, fine for me.)
It's not just this author that's missing. There're also a bunch of others which I simply ignored not spam with a bunch of useless update authors file PRs. It's not relevant for any build stages, so I don't think its worth wasting the time in picking the individual PRs and try to backport them. #10650 (comment) |
Who else is missing? Though actually nobody is missing here. This PR should just need the additional line added by d81607d to be happy: Line 4 in 4e840cf So I don't see a reason why this PR shouldn't just include it. Bypassing merge rules shouldn't become the norm. |
There's also the dependabot user https://github.com/Icinga/icinga2/actions/runs/19765323465/job/56636585443 which requires the workaround in the authors-file workflows too. |
Why isn't that commit part of the PR that actually introduced the naming change? I just didn't wanted to waste another GHA runs for such useless commit but fine I will backport that commit too. |
|
Why did you include the following two PRs as well? I don't see how they are related to this at all. These should have been backported the first there was a failed authors file check on the support/2.14 branch because a dependabot commit was backported instead of force-merging the PR. Can we please do this properly, i.e. now as a separate PR instead of sneaking it into another one? |
yhabteab
force-pushed
the
backport-10278-to-support/2.14
branch
from
2d5287b to
c3ccfa7
Compare
3 participants
Backport of: