Closed takluyver closed 5 years ago
In the older runs, as described in #131, only data files from the fast detectors ended up in proc, so combining raw & proc was useful. In more recent experiments I just found, it looks like the DA
files from the data aggregators are also copied to proc, so maybe this isn't so important any more. But maybe those files are being copied as a workaround to make it easier to process the data from a run together.
Are all DA
files copied to proc? Because some detector data (10Hz, like Jungfrau, Epix, ...) are stored in DA
files. So running calibration on those detector would copy also the specific DA
file.
I thought about that, but it looks like files with no detectors in are also getting copied. See e.g. /gpfs/exfel/exp/SCS/201921/p002409/proc/r0102
This seems like the right thing to do anyway, regardless of what file is copied or not... LGTM
This partly addresses #131: it should make it possible to combine proc detector data with raw data, by deselecting the detector sources in the raw data.