pysat / pysatNASA

pysat support for NASA Instruments
BSD 3-Clause "New" or "Revised" License
21 stars 6 forks source link

ENH: timed see and xarray #136

Closed jklenzing closed 1 year ago

jklenzing commented 2 years ago

Description

Addresses #14

Type of change

How Has This Been Tested?

Load and inspect a day of data.

Test Configuration

Checklist:

jklenzing commented 1 year ago

Noting that this option may be replaced by netCDF files, but #103 needs to be implemented first.

jklenzing commented 1 year ago

Revisiting this issue. I've updated based on the latest develop. Ultimately this is where we want this data product to be, but my initial concern is that cdflib spits out warnings because the original files are not formatted properly, which is well outside our jurisdiction. Sample output when I load:

ISTP Compliance Warning: Epoch is listed as the DEPEND_0 for variable WAVE_SP, but the dimensions do not match.
ISTP Compliance Warning: Epoch is listed as the DEPEND_0 for variable WAVE_LINE, but the dimensions do not match.
ISTP Compliance Warning: Epoch is listed as the DEPEND_0 for variable NAME_LINE, but the dimensions do not match.
ISTP Compliance Warning: Epoch is listed as the DEPEND_0 for variable LOGT_LINE, but the dimensions do not match.
ISTP Compliance Warning: Epoch is listed as the DEPEND_0 for variable WAVE_XPS_1, but the dimensions do not match.
ISTP Compliance Warning: Epoch is listed as the DEPEND_0 for variable WAVE_XPS_2, but the dimensions do not match.
Warning: Variable WAVE_SP listed DEPEND_0 as Epoch, but they have different dimension lengths.
Warning: Variable WAVE_LINE listed DEPEND_0 as Epoch, but they have different dimension lengths.
Warning: Variable NAME_LINE listed DEPEND_0 as Epoch, but they have different dimension lengths.
Warning: Variable LOGT_LINE listed DEPEND_0 as Epoch, but they have different dimension lengths.
Warning: Variable WAVE_XPS_1 listed DEPEND_0 as Epoch, but they have different dimension lengths.
Warning: Variable WAVE_XPS_2 listed DEPEND_0 as Epoch, but they have different dimension lengths.

Options:

@JonathonMSmith , @aburrell, any preferences to which route we take? At this stage I've reached acceptance, but I am not sure if new users will be opening issues based on the warnings.

aburrell commented 1 year ago

My thoughts:

jklenzing commented 1 year ago

Plan is to move forward with code using cdf, with docstrings and comments documenting expected warnings and roadmap in code.