-
-
Notifications
You must be signed in to change notification settings - Fork 43
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Use conda-forge dependencies #8
Use conda-forge dependencies #8
Conversation
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 ( |
Interesting, appears this uses |
So, this is a pretty subtle point, but |
After investigating the |
This is ready to go on my end. |
Did we confirm this fixes the SSL issues? |
Not yet. Though |
--system-libarchive \ | ||
--system-liblzma \ | ||
--system-zlib \ | ||
--no-qt-gui \ |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Please consider adding the following:
--parallel=${CPU_COUNT}
--
-DCMAKE_BUILD_TYPE:STRING=Release \
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The CPU_COUNT
value tends to be inaccurate on the CIs. We have explored setting this through MAKEFLAGS
in PR ( conda-forge/conda-forge-build-setup-feedstock#1 ). However, that turns out to be more trouble than it is worth and removal is thus being proposed in PR ( conda-forge/conda-forge-build-setup-feedstock#18 ). Ideally we will override CPU_COUNT
so that it can be used responsibly, but we will need to get some other players on-board first.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
See this issue ( conda/conda-build#1117 ) for overriding CPU_COUNT
.
Anything I could to to help with this patch? |
If you can come up with a test, for |
I took a stab at it here: blowekamp@c41ba77 HTH |
Thanks @blowekamp. I made some tweaks to it, but have left you as author on that commit. It seems to work ok for me locally. Let's see what the CIs think. |
Seems it passes that test. |
This looks good to me. @blowekamp - are you interested in being added as a maintainer here? |
Yes, I'll continue to help out with this recipe. |
Thanks @blowekamp. Was thinking of asking the same thing. Have added you on. An email should be sent to you later that will ask you to join conda-forge (after this is merged). |
Alright, @blowekamp @patricksnape the Linux and OS X builds of this are out in the wild. Windows builds are pending though there was no actionable change for them with this addition. You should be good to go. Please let me know if you run into any issues. Thanks for raising this and helping me resolve it. 😄 |
@jakirkham Still seeing this error for the OpenCV3 recipe. Edit: Seems I'm not getting this build actually - any idea why I'm pulling build 3 rather than build 4? |
Fixes #7
Tries to use conda-forge dependencies as much as possible instead of trying to use CMake internally built dependencies. This should speed up the build, improve portability, and fix various subtle issues like not having certs for curl.
This adds some dependencies, but some of them like
bzip2
are not here at conda-forge yet. There are more dependencies we should add likeexpat
andxz
, which don't exist on Windows yet.There are some other dependencies like(EDIT: everything is packaged on UNIX and added as a dependency) Finally, we still need to address pinnings here and probably add a test for using SSL. Other questions include whether we should build this with Qt support as that is an option and could be desirable in some cases.jsoncpp
andlibarchive
, which are simply not packaged yet.Feedback welcome on the progress thus far and any of these points that still need to be addressed.
cc @patricksnape @msarahan