Error building XRootD@v5.5.0 due to setuptools

Dear experts,

I’m building O2Physics on macOS Monterey and it fails building XRootD. The log is rather short and the problem seems to be related to the version of setuptools which is looked for:

++ [[ -e /Users/auras/alice/sw/SOURCES/XRootD/v5.5.0/v5.5.0/bindings ]]
+++ /usr/bin/env python3 -c ‘import sys; print(sys.executable)’
++ PYTHON_EXECUTABLE=/opt/homebrew/opt/python@3.10/bin/python3.10
+++ /opt/homebrew/opt/python@3.10/bin/python3.10 -c ‘import sys; print(f"{sys.version_info.major}.{sys.version_info.minor}")’
++ PYTHON_VER=3.10
++ case $ARCHITECTURE in
++ [[ -n ‘’ ]]
+++ brew --prefix openssl@1.1
++ OPENSSL_ROOT=/opt/homebrew/opt/openssl@1.1
+++ brew --prefix
++ CMAKE_FRAMEWORK_PATH=/opt/homebrew/Frameworks
+++ xcrun --show-sdk-path
++ export ‘CFLAGS=-fPIC -O2 -isysroot /Applications/’
++ CFLAGS=‘-fPIC -O2 -isysroot /Applications/’
++ unset UUID_ROOT
+++ python3 -c ‘import setuptools; print(setuptools.version)’
++ ‘[’ 67.3.1 ‘!=’ 60.8.2 ‘]’
++ echo ‘Please install setuptools==60.8.2’
Please install setuptools==60.8.2
++ exit 1

Any idea how to proceed?

Thanks in advance!

Hi @auras ! try with an updated alidist (that have xrootd 5.5.2), it should improve the behavior under macos stuff …

Hi @asevcenc , thanks for the hint, but I still have the same problem with XRootD 5.5.2 :frowning:

well, the only obvious problem is that:

++ ‘[’ 67.3.1 ‘!=’ 60.8.2 ‘]’
++ echo ‘Please install setuptools==60.8.2’
Please install setuptools==60.8.2

so try to run in your user:
python3 -m pip install --user setuptools==60.8.2

moreover, if you can spare the time, make a tryout with commenting in from your alidist
this lines:
and then do a python3 -m pip install --user --upgrade setuptools
and see how it goes (but please, tell us the result :slight_smile: ) … if you obtain positive result, maybe this restriction can be removed from recipe

Thanks @asevcenc, I tried commenting lines 46-49 of and XRootD compiled without any more error (with setuptools-67.3.1). I don’t know if the problem is related, but now the building fails in arrow@v9.0.0-alice1, the log is available here: CERNBox

Hi, the problem with the building of arrows was fixed using the procedure spotted here: