Re: [sage-devel] Re: Sage 9.2 fails to build because of cysignals on macos 10.15.7

2020-12-15 Thread Dima Pasechnik
You have weird stuff in /usr/local, e.g. what's /usr/local/include/stdio.h I see in your log? No wonder it can break things. On Tue, Dec 15, 2020 at 12:40 PM Pierre Vanhove wrote: > > hi > > Since no compilation was succesfull I tried sage-9.3-beta4. > > My homebrew version is > > % brew

Re: [sage-devel] Re: Sage 9.2 fails to build because of cysignals on macos 10.15.7

2020-12-15 Thread Dima Pasechnik
the error is gcc -bundle -undefined dynamic_lookup -L/usr/local/lib -isysroot /Library/Developer/CommandLineTools/SDKs/MacOSX10.15.sdk -L/Users/pierre/local/sage-9.2/local/lib -Wl,-rpath,/Users/pierre/local/sage-9.2/local/lib -Wp,-U_FORTIFY_SOURCE

[sage-devel] Re: Sage 9.2 fails to build because of cysignals on macos 10.15.7

2020-12-14 Thread John H Palmieri
Actually, I can't recommend having Sage build its own Python. I get failures with several packages (gmpy2, markupsafe, cython) of this sort: Successfully built gmpy2 ... ERROR: gmpy2-2.1.0b5-cp38-cp38-macosx_10_9_x86_64.whl is not a supported wheel on this platform. On Monday, December 14,

[sage-devel] Re: Sage 9.2 fails to build because of cysignals on macos 10.15.7

2020-12-14 Thread John H Palmieri
Maybe something is going wrong with using the homebrew version of Python — some conflict somewhere — so could you try make distclean ./configure --with-system-python3=no make (I'm having different problems with the system python: see my posts about MACOSX_DEPLOYMENT_TARGET.) -- John On