boost: update to 1.77.0 (#6872)

[leha-bot]

Specify library name and version: boost/1.77.0

This is also a good place to share with all of us why you are submitting this PR (specially if it is a new addition to ConanCenter): is it a dependency of other libraries you want to package? Are you the author of the library? Thanks!

Bump Boost to 1.77.0. All deps have been updated by Python script rebuild-dependencies.py

The boost-mpi-check.patch has been rebased to 1.77.0 version.

Closes #6872, #7157.

---

• I've read the guidelines for contributing.
• I've followed the PEP8 style guides for Python code in the recipes.
• I've used the latest Conan client version.
• I've tried at least one configuration locally with the
  conan-center hook activated.

[ghost]

I detected other pull requests that are modifying boost/all recipe:

• - [boost] fix libiconv detection (needed by boost locale) for iOS, Android #7146
• [boost] Apply patch to fix https://github.com/boostorg/graph/issues/268 #6132
• [boost] For Visual Studio, pass some values of settings.compiler.toolset to B2 #5119
• boost: use clang-win toolset when compiler=Visual Studio and compiler.toolset=ClangCL #4749

This message is automatically generated by https://github.com/ericLemanissier/conan-center-conflicting-prs so don't hesitate to report issues/improvements there.

[leha-bot]

@jgsogo sorry for another mention; could you help with this weird CI error? 😊 The MacOS build has strangely failed (at first and second time)

[gummif]

Maybe related to opencv/opencv#14146, but have no idea where to apply that fix.

Maybe add to the test package CMakeLists (see https://skia.googlesource.com/external/github.com/libjpeg-turbo/libjpeg-turbo/+/refs/tags/1.5.90%5E%21/)

     if(APPLE)
       if(NOT CMAKE_SHARED_LIBRARY_RUNTIME_C_FLAG)
         set(CMAKE_SHARED_LIBRARY_RUNTIME_C_FLAG "-Wl,-rpath,")
       endif()
     endif()

[leha-bot]

Maybe related to opencv/opencv#14146, but have no idea where to apply that fix.

Maybe add to the test package CMakeLists (see https://skia.googlesource.com/external/github.com/libjpeg-turbo/libjpeg-turbo/+/refs/tags/1.5.90%5E%21/)

     if(APPLE)
       if(NOT CMAKE_SHARED_LIBRARY_RUNTIME_C_FLAG)
         set(CMAKE_SHARED_LIBRARY_RUNTIME_C_FLAG "-Wl,-rpath,")
       endif()
     endif()

It looks like the test_package's build failure, as the Boost is using Boost.Build (b2) for building yourself.
We could set this policy (CMP0068) in test_package's CMakeLists.txt, thanks for the idea 👍

It also seems that the MacOS image also uses the old CMake ( < 3.9)...

[prince-chrismc]

so the new dependencies-xxxxx.yml file is same as was for 1.76.0.

Did you run https://github.com/conan-io/conan-center-index/blob/365086d2c4164b27f6e89eb07fc165581d031527/recipes/boost/all/rebuild-dependencies.py?

[prince-chrismc]

😖 Why are windows and macos coming as duplicate package ids?

-- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | -- | --
DUPLICATED | boost/1.72.0 | 524ea35a8120baabdde02483add58d81bf541327 | Windows | Windows | x86_64 | x86_64 | Visual Studio | 14 | MT |   | Release |   | True |   |   |  
DUPLICATED | boost/1.72.0 | 524ea35a8120baabdde02483add58d81bf541327 | Macos | Macos | x86_64 | x86_64 | apple-clang | 10.0 |   | libc++ | Release |   | True  |   |  

//cc @madebr (our boost guru)

[madebr]

Thanks for the nomination @prince-chrismc.
The duplicated builds are no problem. They are the header-only boost version for Macos/Linux/Windows.

It looks like boost/1.72 is failing in the test package for arch=armv8 os=Macos.

CMake Error at build/9e9e53fc2b762590855e996df53de3efb6fd7fb9/Findlibbacktrace.cmake:39 (add_library):
  Attempting to use @rpath without CMAKE_SHARED_LIBRARY_RUNTIME_C_FLAG being
  set.  This could be because you are using a Mac OS X version less than 10.5
  or because CMake's platform configuration is corrupt.
Call Stack (most recent call first):
  build/9e9e53fc2b762590855e996df53de3efb6fd7fb9/FindBoost.cmake:1599 (conan_package_library_targets)
  CMakeLists.txt:20 (find_package)

I nominate @SpaceIm as he's our Apple go to guy.
All the errors have to do with libbacktrace. Perhaps something is wrong with that package?

[dmn-star]

My iOS/Catalyst/Android profiles have always looked like this.
...
boost:without_locale = True
boost:without_log = True
boost:without_stacktrace = True

I know that it is not a solution for cci, but I have never been able to build the boost with local/log/stacktrace.

[dmn-star]

https://github.com/ianlancetaylor/libbacktrace
I couldn't see at a quick glance whether that supports ARM64 or not. Maybe disable the support for Arm64 (iOS/Catalyst/M1/Android) like for Windows. From the boost recipe

stacktrace_backtrace not supported on Windows
if self.settings.os == "Windows":
del self.options.with_stacktrace_backtrace

[SSE4]

https://github.com/ianlancetaylor/libbacktrace
I couldn't see at a quick glance whether that supports ARM64 or not. Maybe disable the support for Arm64 (iOS/Catalyst/M1/Android) like for Windows. From the boost recipe

it seems like ARM64 is supported in code to some degree: ianlancetaylor/libbacktrace@4e548e7#diff-76ce24fe823ff3c52a40864352cbd78278e3dbded9e905904f266feaf5aeca8aR760
I'll check it tomorrow and reply back

[SpaceIm]

I nominate @SpaceIm as he's our Apple go to guy.

Nice try, but I don't know 😄

[madebr]

I nominate @SpaceIm as he's our Apple go to guy.

Nice try, but I don't know 😄

Detective @SSE4 🕵️ is on the case.

[dmn-star]

https://github.com/ianlancetaylor/libbacktrace
I couldn't see at a quick glance whether that supports ARM64 or not. Maybe disable the support for Arm64 (iOS/Catalyst/M1/Android) like for Windows. From the boost recipe

it seems like ARM64 is supported in code to some degree: ianlancetaylor/libbacktrace@4e548e7#diff-76ce24fe823ff3c52a40864352cbd78278e3dbded9e905904f266feaf5aeca8aR760
I'll check it tomorrow and reply back

Do you have access to M1 Mac to verify this?

ps: google says swift-server/swift-backtrace#19

@leha-bot If you want to close multiple issues at once use full syntax for each issue
Closes #6872, closes #7157

[SSE4]

Do you have access to M1 Mac to verify this?

unfortunately I don't have access. at least, it cross-compiles successfully, but I can't run any tests.
also, there are aarch64 packages for libbacktrace in several distros, so I believe it works.

[conan-center-bot]

All green in build 9 (8a5cd5a4ebfea91da01f9563022029ec967334fa):

• boost/1.70.0@:
  All packages built successfully! (All logs)

• boost/1.69.0@:
  All packages built successfully! (All logs)

• boost/1.71.0@:
  All packages built successfully! (All logs)

• boost/1.75.0@:
  All packages built successfully! (All logs)

• boost/1.72.0@:
  All packages built successfully! (All logs)

• boost/1.74.0@:
  All packages built successfully! (All logs)

• boost/1.73.0@:
  All packages built successfully! (All logs)

• boost/1.76.0@:
  All packages built successfully! (All logs)

• boost/1.77.0@:
  All packages built successfully! (All logs)

[madebr]

Why did c3i build all combinations?
Isn't building only 1.77.0 enough?

[SSE4]

Why did c3i build all combinations?
Isn't building only 1.77.0 enough?

good question... recipe has:

exports_sources = ['patches/*']

and commit adds a new patch. even if it's not used by the old versions, I guess it's still exported and considered as a part of recipe revision?

[madebr]

That makes sense.