Skip to content

Conversation

dawsoneliasen
Copy link
Contributor

No description provided.

Copy link
Owner

@bangerth bangerth left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Nice to see that you got it to work!

- name: install dependencies
run: |
brew install cmake boost doxygen
git clone https://github.com/tjhei/numdiff.git
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

That's the repository of my friend Timo Heister, but it's not the master repository for the numdiff project. Is that what you had in mind?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Oh, that's weird - it's the first thing that comes up when I search for numdiff. Can you point me to the master for the numdiff project?

Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

The project lives here: https://www.nongnu.org/numdiff/

You should be able to get it via wget from http://download-mirror.savannah.gnu.org/releases/numdiff/ as a .tar.gz file that you can unpack via tar xvzf file.tar.gz.

CMakeLists.txt Outdated
# Set paths correctly to take BOOST from wherever it is
# installed on the current system
FIND_PACKAGE(Boost 1.39 REQUIRED)
FIND_PACKAGE(Boost REQUIRED COMPONENTS system)
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Was there a particular reason to remove the minimum version as a requirement?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Still trying to figure it out, windows cmake fails saying it could not find boost. I found a stack overflow post suggesting this might help, but it's still failing...

Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Let's try to figure that out on Wed.

Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Try undoing this change.

@bangerth
Copy link
Owner

If you look at the current version, BOOST_DIR is apparently not set: the dir call just outputs the content of the current directory, not of the BOOST_DIR directory (which is apparently an empty string). Is setx the right way to set environment variables in the powershell?

@dawsoneliasen
Copy link
Contributor Author

If you look at the current version, BOOST_DIR is apparently not set: the dir call just outputs the content of the current directory, not of the BOOST_DIR directory (which is apparently an empty string). Is setx the right way to set environment variables in the powershell?

Yep, I think you're right. Let me try something else.

@dawsoneliasen
Copy link
Contributor Author

@bangerth Any thoughts at this point? The environment variables appear to be set up correctly, but the error message persists. I have looked around for solutions but I haven't found anything conclusive. Here is a windows setup guide from the boost docs. Some SO conversations seems suggest that an additional install is required, e.g. sudo apt-get install libboost-all-dev, but I'm not sure what the windows equivalent of this would be.

@bangerth
Copy link
Owner

I'm colossally ignorant about Windows, so really don't know what to make of the error.

That said, cmake puts output from its failing checks into CMakeFiles/CMakeError.log. What would be interesting to do is output the contents of this file somehow. On Unix, one would do this in the following way:

  cmake .  ||  cat CMakeFiles/CMakeError.log

where the || ensures that if the first command fails, the second command is still executed (necessary because false || something might still yield true; on the other hand, if the first command succeeds, there is no need to even execute the second command -- it's a funny way to write things, but widely used).

I don't know whether there is a corresponding way on the Windows shell. The alternative is to use the "artifacts" system of CI where certain output files are uploaded I don't know off-hand how this is set up, but I'm sure you can find it online -- the file CMakeFiles/CMakeError.log (and maybe also CMakeFiles/CMakeOutput.log) should then be uploaded as an artifact.

@dawsoneliasen
Copy link
Contributor Author

Upload Mac output files as artifacts to see what's going wrong.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants