Update pin for boost

[regro-cf-autotick-bot]

This PR has been triggered in an effort to update the pin for boost. The current pinned version is 1.74.0, the latest available version is 1.76.0 and the max pin pattern is x.x.x. This migration will impact 78 feedstocks.

Checklist:

• The new version is a stable supported pin.
• I checked that the ABI changed from 1.74.0 to 1.76.0.

**Please note that if you close this PR we presume that the new pin has been rejected.

@conda-forge-admin please ping boost
This PR was generated by https://github.com/regro/autotick-bot/actions/runs/808040834, please use this URL for debugging

[conda-forge-linter]

Hi! This is the friendly automated conda-forge-webservice.

I was asked to ping @conda-forge/boost and so here I am doing that.

[conda-forge-linter]

Hi! This is the friendly automated conda-forge-linting service.

I just wanted to let you know that I linted all conda-recipes in your PR (recipe) and found it was in an excellent condition.

[chrisburr]

Do we want to change macOS to use C++ 17 by default as part of this?

See: conda-forge/clang-compiler-activation-feedstock#17

[xhochy]

Last time we had agreed on doing the boost migration for old and new version at the same time. I have updated it accordingly.

Reading conda-forge/clang-compiler-activation-feedstock#17 (comment) I'm not sure whether this would work and whether we failed to do the right thing in the boost update as we didn't increase the C++ standard over there.

[chrisburr]

I'm not sure whether this would work and whether we failed to do the right thing in the boost update

We could mark the _0 macOS build as broken rather than waiting 6 months. It hasn't been downloaded much so I don't think it would cause much trouble.

[jakirkham]

As noted in the other thread, added the C++17 issue to the agenda for the meeting tomorrow

[h-vetinari]

conda-forge/clang-compiler-activation-feedstock#75 is in, so this should be unblocked now?

[beckermr]

Shall we jump ahead to 1.78 now?

[jakirkham]

The feedstock is on 1.79.0. Would suggest going to that given there are other fixes that have been made to that and not prior releases

[beckermr]

We only migrate even versions and have been only doing that for a while.

[beckermr]

boost migrations tend to be very slow, reveal tons of breaks and bugs, and require lots of person power.

[jakirkham]

Understand the need to do fewer migrations. Hence suggesting to go ahead with the newest version

Indeed we have had a convention of using even versions. However given it has been a while since we have updated Boost and we are planning to jump ahead. Would push for latest with needed fixes over following a convention that might not be best serving us in this instance

[h-vetinari]

Would push for latest with needed fixes over following a convention that might not be best serving us in this instance

Agreed

(edit 1.80 is almost out, so it's not like we'd be jumping to the very cutting edge with 1.78...)

[jakirkham]

Do you know when 1.80.0 is due?

[h-vetinari]

Do you know when 1.80.0 is due?

Based on the patterns of previous releases in about 2 weeks. But I wouldn't go to 1.80 directly (basically no upstream will have tested compat yet, exacerbating the problems Matthew mentioned) - IMO 1.78 is safer.

[izahn]

I think going to 1.78 makes sense. Is there anything blocking us from doing that in #2757 ?

[izahn]

Any updates here @conda-forge/boost ? We need #2757 for rstudio updates.

[izahn]

@conda-forge/core I would really appreciate an update here, we need boost>=1.78 for rstudio-feedstock updates.

[h-vetinari]

There's now a bot-PR for 1.80: #3268

Given the concerns raised by Matthew (and the size of the bump 1.74->1.80 would be), I still think we should go with 1.78 in #2757.

[izahn]

It looks like there is some consensus around going with 1.78. Can we go ahead and merge #2757 ?

[beckermr]

closing since we went with 1.78.0