Date: Thu, 12 Sep 2019 23:51:28 +0000 From: bugzilla-noreply@freebsd.org To: python@FreeBSD.org Subject: [Bug 240476] security/py-fido2: Add FreeBSD support Message-ID: <bug-240476-21822-yWgdcp5PC2@https.bugs.freebsd.org/bugzilla/> In-Reply-To: <bug-240476-21822@https.bugs.freebsd.org/bugzilla/> References: <bug-240476-21822@https.bugs.freebsd.org/bugzilla/>
next in thread | previous in thread | raw e-mail | index | archive | help
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=3D240476 --- Comment #13 from Michael Gmelin <grembo@FreeBSD.org> --- (In reply to Kubilay Kocak from comment #1) Thanks for your response and the clear instructions Kubilay. > - Confirm the changes pass QA (portlint, poudriere) Confirmed, please find attached the output of `poudriere testport security/py-fido2' runs for various combinations of releases and architectu= res. portlint -a is happy (output also attached). > - If there's a test suite for fido2, confirm it passes with this patch (a= nd adding TEST_DEPENDS / test: target support to port) The test suite didn't work, as it would unconditionally import all test sources, including the linux tests, which require pyfakefs, which isn't available in ports (I tried a quick port, but it's quite linux specific and would've required even more work, just to support a unit test that won't run anyway). To fix this, setup.py now uses pytest, which allows exclusion of f= iles in conftest.py (also gets rid of "you're not on darwin" warnings). Test target support is now in the port skeleton, so `make test' does the Ri= ght Thing(tm). I also added unit tests for the FreeBSD specific bits and ran `m= ake test' in all the poudriere combinations mentioned above manually (using a s= hell after invoking `poudriere testport -i'). > - Submit the patches in an upstream PR and add the PR title/URL as a comm= ent in the patch header To to so, some more work was necessary (like adding FreeBSD unit tests, changing readme, etc.). You can find the pull request here https://github.com/Yubico/python-fido2/pull/64 > Also, could you register/publish uhid-freebsd (under that name) in PyPI, = so that upstream can depend on it in setup.py:install_requires using an env= ironment marker ; sys_platform =3D=3D 'freebsd' PyPI account created and package published (https://pypi.org/project/uhid-freebsd/). devel/py-freebsd-uhid now pulls f= rom there instead of github. Anything else? --=20 You are receiving this mail because: You are the assignee for the bug.=
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?bug-240476-21822-yWgdcp5PC2>