Handle new z-serie.

context

At tomo-bliss side a z-serie has been redefined as:

Previously it use to be

Today (1.0.8) if a new z-series it can contains darks / flat only at the beginning or at the end. As a consequence the converted NXtomo will not contain dark/flat (except the one at the beginning or at the end)

Furthermore it use to be in the same sequence when now it is split into n sequences... As a consequence we can handle dark / flat copy when the full file is converted. But when only a subset of the files is selected (using entries / sub_entries_to_ignore) we cannot handle them. Nevertheless this is probably a corner case.

details

• requires tomoscan!226 (merged) to be merge first
• add processing
• test on real (new) dataset
• test on real (old) dataset
• add unit test.
  • warning: I upgraded the test layout (moving to pytest). The important part on nxtomomill.converter.hdf5.test.test_hdf5converter is the addition of test_z_series_dark_flat_copy which specifically test the dark / flat copy for z-series-v3

note: z-series provided by Clemence is available in /scisoft/tomo_data/z_series/z_series_v3

extra

• script used to edit the proposal file: clean_proposal.py
• we will need to forward it to the main branch. Having to 1.0 branch will allow us to create a release (1.0.9) and a deployment to have this feature before the next release.