Bug: ERA5 hourly data (skipped files)

[ADRIE-A3]

Describe the bug
ERA5 hourly data are not read on the hortense machine, only ERA5 daily and monthly data. in general a lot of files are skipped when using the inputmanager (vp.InputManager(machine='hortense') )

Expected behavior
To be able to read ERA5 hourly data.

Environment
(super) computer - hortense
Branch - main?
Conda environment - '/dodrio/scratch/projects/2022_205/softwares/vsc46032/conda_envs/valenspy_dev'
Other relevant environment information:

Minimal Reproducible Example
see notebook /dodrio/scratch/projects/2022_200/project_output/RMIB-UGent/vsc45381_leon/VALaro/src/analyse_eurocities.ipynb

Additional context

[kobebryant432]

This is currently like this by design:

  ERA5: #Note that there is an aggregation variable here! What to do with this? This is messy as it can not be uniquely identified by the filename!
    root: /dodrio/scratch/projects/2022_200/external/era5/
    pattern: <domain_id>/<variable_id>/<frequency>/<variable_id_2>_era5_<domain_id_2>_<frequency_2>_<aggregation>_<time_period>.nc
    meta_data:
      source_id: "ERA5"
      source_type: "reanalysis"
      experiment_id: "reanalysis"
      version: "ERA5.0"
      resolution: "0.25°"
      institution_id: "ECMWF"
      time_format: "%Y"   

This yaml is used and the pattern in particular the "aggregation" section which is not defined for ERA5 hourly data. When matching for example the file:

/dodrio/scratch/projects/2022_200/external/era5/europe/2m_temperature/hourly/2m_temperature_era5_europe_hourly_1941.nc it does not match the provided pattern. This is because _aggregation is not found.

There are 2 solution:

1. Add some functionality in Valenspy Feature: Add option to input yml file to specify multiple patterns #273 (ongoing) - should be an easy fix
2. Quick fix, similar to adapt CCLM inputconvertor identifiers #271

@ADRIE-A3

[kobebryant432]

Fixed with #271, see #273