Jon Niehof
Jon Niehof
Right now a simple `import spacepy` brings in datamodel, time, and toolbox. We're force-importing the datamodel, which sort of makes sense, but the others get to be too much--for instance,...
This is a big one but SpacePy docstrings really should have notation on when a function was added to SpacePy. Since we're well on our way to 0.2.2 I don't...
`plot.utils.add_arrows` could benefit from a cool example. I'm sure @spacecataz has something that would be relevant. I can help get it worked up into the docs, just need something to...
If the FILLVAL attribute doesn't match the expected for EPOCH16 vars, attempting to build the error message gives: `TypeError: tuple indices must be integers, not tuple` The conversion of the...
CDF variables shouldn't have nan (that's what FILLVAL is for.) validrange raises a warning if there are: ``` site-packages/spacepy/pycdf/istp.py:519: RuntimeWarning: invalid value encountered in less idx = (data < attrval)...
Ticktock uses Julian dates up through 1582-10-4 and Gregorian from 1582-10-15 on (inclusive). But it doesn't reject ISO or UTC inputs in the 1582-10-5 through 1582-10-14 range; conversions to (say)...
This is really only checking that logical_source matches the beginning of the filename..for a file `psp_isois-epilo_l2-ic_20200101_v1.cdf` this would accept `psp_`, `psp_isois-epilo_l2-ic_`, and `psp_isois-epilo_l2-ic_20200101_v1.cdf`, all of which are invalid.
The pycdf.lib.cdftypenames dict needs to be fully documented, and in particular linked from functions that return types.
The variable checks should verify that CATDESC isn't just the variable name, which is a pretty common way to go and nearly always wrong.
pycdf ISTP variable checks require that the variable named in DELTA_PLUS_VAR/DELTA_MINUS_VAR match the type of the "parent" variable. For Epoch types this is a bit weird, as e.g. a one...