At the moment, the GONG map used for the PFSS extension is obtained with a hard-coded Timedelta of one hour:
|
desired_time = pd.to_datetime(timestr, yearfirst=True) |
|
desired_time_plus_hour = desired_time + pd.Timedelta(hours=1) |
|
|
|
result = Fido.search(attrs.Time(desired_time, desired_time_plus_hour), attrs.Instrument('GONG')) |
This works fine when there are hourly GONG maps (e.g.,
https://gong2.nso.edu/oQR/zqs/202312/mrzqs231205/), but runs into problems when there are fewer maps available (e.g.,
https://gong2.nso.edu/oQR/zqs/200612/mrzqs061205/).
This should be better handled. Not sure if automatically a "father away" (in time) map should be used, or the user should be informed about the available maps so that they can make the decision, e.g., to change the time for the SolarMACH object accordingly.
(Partly connected to #61)
At the moment, the GONG map used for the PFSS extension is obtained with a hard-coded
Timedeltaof one hour:solarmach/solarmach/pfss_utilities.py
Lines 648 to 651 in 4e110cb
This works fine when there are hourly GONG maps (e.g., https://gong2.nso.edu/oQR/zqs/202312/mrzqs231205/), but runs into problems when there are fewer maps available (e.g., https://gong2.nso.edu/oQR/zqs/200612/mrzqs061205/).
This should be better handled. Not sure if automatically a "father away" (in time) map should be used, or the user should be informed about the available maps so that they can make the decision, e.g., to change the time for the
SolarMACHobject accordingly.(Partly connected to #61)