Panu Lahtinen

The SEVIRI datasets have these time-related keys: ``` nominal_start_time 2023-05-10 08:00:10.708720 nominal_end_time 2023-05-10 08:05:10.430222 start_time 2023-05-10 08:00:10.708000 end_time 2023-05-10 08:04:17.969000 ``` There doesn't seem to be exact times available...

Ideally we'd implement the download and data handling to `fmiopendata`. It shouldn't be very complicated, but I'm not sure when I'll find the time to implement this. In the meanwhile,...

@mraspaud yes, indeed. I'm planning on working on this and the performance tweaking during PCW.

I've made some progress based on https://github.com/pytroll/satpy/pull/2237/commits/bb7c313aff078fe0ae3e0618f78b913a310bbfe5 . With that, I can create 20 resampled day/night composites in 140 seconds, which is roughly 40 seconds more than by using the...

Closing due to #2305 which is based on recently merged PRs handling FCI data.

The tests seem to be failing, lots of cases where the error is something like `AttributeError: module 'satpy.readers' has no attribute 'yaml_reader'`. See https://github.com/pytroll/satpy/actions/runs/3712880243/jobs/6294915398#step:10:22319 for the full list. With a...

I get the same failures locally.

This passes: ``` pytest satpy/tests/reader_tests/ ``` but this fails ``` pytest satpy/tests/test_readers.py satpy/tests/reader_tests/ ``` So something is cached in `test_readers.py` and not cleared?

The *wavelength* units look wrong. In the error message of the load it is `unit='µm'` and in the `ncdump` output `string IR_108:wavelength = "10.8 µm (9.8-11.8 µm)" ;`.

Tried with some SEVIRI HRIT data I have: ``` In [9]: scn['IR_108'].attrs['wavelength'] Out[9]: WavelengthRange(min=9.8, central=10.8, max=11.8, unit='µm') ``` And after saving `ncdump -h` shows ``` string IR_108:wavelength = "10.8 µm ...