use cmake to obtain correct soname

conda-forge / libarchive-feedstock

A conda-smithy repository for libarchive.

BSD 3-Clause "New" or "Revised" License

2 stars 26 forks source link

use cmake to obtain correct soname #70

Closed jaimergp closed 10 months ago

jaimergp commented 2 years ago

Checklist

[X] Used a personal fork of the feedstock to propose changes
[ ] Bumped the build number (if the version is unchanged)
[ ] Reset the build number to 0 (if the version changed)
[ ] Re-rendered with the latest conda-smithy (Use the phrase code>@<space/conda-forge-admin, please rerender in a comment in this PR for automated rerendering)
[ ] Ensured the license file is being packaged.

See #69 for more details

conda-forge-linter commented 2 years ago

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.

jaimergp commented 2 years ago

@conda-forge-admin, please rerender

conda-forge-webservices[bot] commented 10 months ago

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.

jaimergp commented 10 months ago

@ocefpaf, this is probably going to become a bigger issue when libmamba becomes default, and I am seeing reports of the defaults<->conda-forge incompatibility in more and more places:

https://github.com/SciTools/workflows/pull/3
https://github.com/conda/conda/issues/13041
https://github.com/conda/conda-libmamba-solver/issues/244
https://github.com/mamba-org/mamba/issues/1826 (this has a handful!)
... and probably more

As a result I'd like to move forward with this PR. I've added symlinks to ensure backwards compatibility with the previous builds and a "release bomb" for 3.8 so we remove them. We'll also need to backport this to the 3.6 and 3.5 branches.

If it doesn't work, well, I'll mark them as broken and go back to the drawing board 😬

JeanChristopheMorinPerso commented 10 months ago

Hi @jaimergp I did a quick check with defaults and I see that we build with -DENABLE_OPENSSL=TRUE on unix systems (except for osx-64 where we don't build with OpenSSL, but I don't know why). Is it something that should be replicated here too? See https://github.com/AnacondaRecipes/libarchive-feedstock/blob/master/recipe/build.sh#L12-L16.

jaimergp commented 10 months ago

@JeanChristopheMorinPerso - True, good catch! I missed that line in configure. Added it now. Thanks!

JeanChristopheMorinPerso commented 10 months ago

Np. But you know what, it's actually the default. I didn't realize it is the default. Anyway, it doesn't hurt to be explicit! Thanks for making the modification!

jaimergp commented 10 months ago

Hm, @isuruf pointed me to this list of items to check when making this precise change (autotools -> CMake), and it might be trickier than expected (or not possible at all). I'll report later.

jaimergp commented 10 months ago

Closing here as we won't be able to fix it this way. Check the issue for more details on the strategy going forward.