On Tuesday, 26 July 2022 at 15:58:18 UTC-7 Matthias Koeppe wrote: > > Whether the method described in section > https://doc.sagemath.org/html/en/installation/conda.html#using-conda-to-provide-all-dependencies-for-the-sage-library-experimental > > works always depends on the conda package maintainers providing packages > that are up to date with respect to our requirements. It's not directly > under control of our project. > > Hm ... so I guess "configure" with these "--with-system-*=true/false/force" can then determine per-prerequisite whether to try to get it from the system or build it by itself? And that the advantage of conda is that it can distribute binaries, that are then hopefully built with approrpriate flags for your particular system? And that the hope is that conda-provided prereqs are a better match than what the OS may provide and/or are more stable (mainly because one can afford to not update a conda env regularly).
Is it possible for prerequisites that conda fails at, such as primesieve and primecount, to flick the "with-system" switches the other way, so that sage builds that by itself? Is there a way to tell the system to get from conda what it can and otherwise build it by itself? It seems like that's what configure does normally (and with a WAY shorter configure command!), so would it be able to do that within conda? It would seem that after: $ conda env update --file environment.yml -n sage-build (which is I think one of the commands where using mamba instead saves gobs of time!!!) conda has already exerted its best efforts to satisfy the prereqs, so anything that now fails should be built by sage itself ... -- You received this message because you are subscribed to the Google Groups "sage-devel" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-devel+unsubscr...@googlegroups.com. To view this discussion on the web visit https://groups.google.com/d/msgid/sage-devel/106b3d56-51e7-44ff-9761-8e1c40a1f88an%40googlegroups.com.
