Gabe Brammer

Thanks @chriswillott . I pushed a fix in https://github.com/gbrammer/grizli/pull/208, which is available in the master branch but not yet in the pip release.

This is now in the release, closing.

This should probably just be `bits=2+4+32768` and keep the mask for `DO_NOT_USE (1)`. Or more generally, I think all of the JWST bits are "warnings" other than `DO_NOT_USE (1)`. And...

This fix seems fine and relatively harmless, but it could be more automatic to also have [exposure_oneoverf_correction](https://github.com/gbrammer/grizli/blob/7da640e68a83f93e1f58802c1553ea33d7184233/grizli/jwst_utils.py#L926) first check for the presence of the `ONEFEXP` keyword in the file and...

I like the modernization idea, but the wholesale changes to the new `Polynomial` behavior are a bit risky, particularly in the way the order of the polynomial coefficients are interpreted...

Thanks for keeping an eye on this. Can you update minimum versions on the `numpy` and `scipy` dependencies in [setup.cfg](https://github.com/gbrammer/grizli/blob/master/setup.cfg) to ensure compatibility with these changes (e.g., `cumtrapz` > `cumulative_trapezoid`)?...

Thanks. Can you find the minimum scipy needed for cumulative_trapezoid?

Merged! Thanks again, @TheSkyentist.

I think I'd prefer that the user have the responsibility for assigning the direct image reference for these cases. You could do this either by simply making a symbolic link...

That can happen if there are spectra that "fall off" the nominal pixel grid of a particular exposure. Sacrificing generality for a bit of speed, the array index bit that...