tomoscan issueshttps://gitlab.esrf.fr/tomotools/tomoscan/-/issues2021-04-06T14:06:13+02:00https://gitlab.esrf.fr/tomotools/tomoscan/-/issues/20add start time and end time getter2021-04-06T14:06:13+02:00paynoadd start time and end time getterThe NXTomo generated by nxtomomill are now saving start and end time. We should provide API to access it (https://gitlab.esrf.fr/tomotools/nxtomomill/-/merge_requests/45).The NXTomo generated by nxtomomill are now saving start and end time. We should provide API to access it (https://gitlab.esrf.fr/tomotools/nxtomomill/-/merge_requests/45).0.5https://gitlab.esrf.fr/tomotools/tomoscan/-/issues/19instanciate __str__ ?2021-04-20T07:55:12+02:00paynoinstanciate __str__ ?For now tomwer has is own 'string' representation of a scan. It could be interesting to generalize it through projects in order to have some homogeneity. In this case it should be done in tomoscan.
We could have `__str__` and `short_str...For now tomwer has is own 'string' representation of a scan. It could be interesting to generalize it through projects in order to have some homogeneity. In this case it should be done in tomoscan.
We could have `__str__` and `short_str` for example. The first one return the full path + entry if necessary. The second one return the os.path.basename for edf or the file name for hdf5 ?0.5https://gitlab.esrf.fr/tomotools/tomoscan/-/issues/13HDF5 - manage z series2021-04-06T14:07:05+02:00paynoHDF5 - manage z seriesIn the case of z-series we will receive one NXTomo - HDF5 containing a sequence of n sub-sequence, each with x frames.
Each sequence will have the same z value (which won't be the case for helicoidal / helical scan).
For now tomoscan ca...In the case of z-series we will receive one NXTomo - HDF5 containing a sequence of n sub-sequence, each with x frames.
Each sequence will have the same z value (which won't be the case for helicoidal / helical scan).
For now tomoscan can only produce one TomoScanBase for the entire sequence.
Nabu abd tomwer will need to 'split' this sequence into several sub-sequence' - one per z.
In order to do this for now we see two approaches:
- HDF5TomoScan can take a 'z' parameter to filter the projection to take into account. And scan factory can automatically try to split the input if the NXTomo looks like a zserie (looking at x, y and z values, knowledge about acquisition type...)
- add a function 'split_on_z(f)' to HDF5TomoScan which will also generate sub HDF5TomoScan(s). Less conveniant but maybe the API fit more to the data ?0.5