It seems that the issue you're encountering is related to the compiler not being able to locate the spglib/spglib.h header file. Even though you have installed spglib via pip, it's likely that the compiler is unable to find the relevant headers in the default include paths. Here are some steps you can take to resolve the issue: 1. Verify Spglib Installation: First, make sure that spglib is installed correctly in your environment and is available for both Python and C/C++ code. Run the following command in your Python environment to check if the spglib module is correctly installed: pip show spglib This will show the details of the installed package, including its installation path. 2. Locate the Header File: You need to locate where spglib/spglib.h is installed. In your Anaconda environment, the header should typically be in a folder like: /home/makyo/anaconda3/include/spglib/ You can search for the file by running: find /home/makyo/anaconda3 -name spglib.h 3. Set the Include Path Manually: If the header file is located in a non-standard directory, you may need to tell the compiler where to find it by adding the correct include path. In your compilation command, you can add an -I flag to specify the include directory: gcc -I/home/makyo/anaconda3/include thirdorder_core.c -o thirdorder Alternatively, you can modify the environment variable CFLAGS to include the correct directory: export CFLAGS="-I/home/makyo/anaconda3/include" Then proceed with the compilation. 4. Install Spglib via Conda (if necessary): If manually setting the include path doesn't work, or if you want a more seamless installation, you can try installing spglib via Conda.

This can help with managing dependencies in a more straightforward way for your environment: conda install -c conda-forge spglib This will ensure that the library is installed with the correct headers and libraries in the expected paths for your Anaconda environment. 5. Check for Deprecated API Warning: While the deprecated NumPy API warning is not directly related to the compilation failure, it would be good to address it by disabling deprecated API usage. You can add the following definition before including any NumPy headers in the code: #define NPY_NO_DEPRECATED_API NPY_1_7_API_VERSION This will prevent the compiler from issuing warnings related to the deprecated API. 6. Rebuild the Package: After addressing the spglib header issue, try rebuilding the package. If you are using pip to install thirdorder, you can run: pip install --no-cache-dir --upgrade thirdorder By following these steps, you should be able to resolve the issue related to the missing spglib/spglib.h file and proceed with the compilation of thirdorder for ShengBTE.

Thanks for your answer. I created a folder named spglib in /usr/include/ and after copying spglib.h to the spglib folder the problem was fixed