Edoardo Pasca

@AnderBiguri could you try to add the following after the [`find_package`](https://github.com/CCPPETMR/SIRF-SuperBuild/blob/master/SuperBuild/External_ACE.cmake#L79) call please? ``` if (NOT ACE_FOUND) message(FATAL_ERROR "ACE not found!") endif() ```

see also #156

Discussion with @KrisThielemans Possible solutions: - [ ] we ask ubuntu to package the `hdf5-config.cmake` - [ ] we ask CMake to attempt to test for the `curl` linking -...

I tested the docker build on master: ``` docker build . --build-arg NUM_PARALLEL_BUILDS=1 --build-arg RUN_CTEST=0 --target=build --no-cache ``` `hdf5-helpers` seems installed however the build fails with the default `USE_SYSTEM_HDF5=ON`

That's what CMake finds. I think the problem is that HDF5 is installed both by conda and on ubuntu system if `BUILD_CIL=ON` (via the dependency on `h5py`). CMake does not...

I tried not to install the `hdf5` packages below https://github.com/SyneRBI/SIRF-SuperBuild/blob/5f1a7f498bf6f41551c19eb53798b5d33af43eec/docker/build_gadgetron-ubuntu.sh#L8-L13 However `libarmadillo-dev` depends on a whole lot of hdf5 libraries and ancillary hdf5 packages. I suggest to move the installation...

On the VM we use the system python, while on docker we install conda or better conda is installed by the base image to run jupyter. I agree that it...

The `HDF5` version on Ubuntu 22.04 is [1.10.7](https://launchpad.net/ubuntu/jammy/+package/hdf5-tools). The versions available on [`conda-forge`](https://anaconda.org/conda-forge/hdf5/files) do not have that!

Looking deeper in the issue, our base image [`quay.io/jupyter/scipy-notebook:ubuntu-22.04`](https://github.com/jupyter/docker-stacks/blob/449ed285b24cfd5642a7537cae144225b5bd03c1/images/scipy-notebook/Dockerfile#L40) does already contain both `h5py` and `hdf5`. ``` # Name Version Build Channel hdf5 1.14.3 nompi_hdf9ad27_105 conda-forge h5py 3.11.0 nompi_py311h439e445_102 conda-forge...

I think this is what I generally do, anyway