Allow fiber loss calculations to be made in xy space rather than radius & azimuth space.

schlafly commented 5 years ago

The primary goal of this PR is to allow offsets to be specificed to GalsimFiberlossCalculator in x & y space rather than radius and azimuth space.

While addressing this, I made a few other minor changes:

removed pile up of COMMENT header cards from saved image files
updated docstrings to reflect radial & azimuthal coordinates rather than x & y

NB: astropy's 'COMMENT' keyword is a bit weird. "overwriting" a comment keyword just adds a new header keyword, preserving past comments. So before the explicit del's were added there was an increasing number of COMMENTs in each image.

@tolgayapici, you'll either want this change or to compute your own "rotated" xy offsets when passing to GalsimFiberlossCalculator.calculate(...).

weaverba137 commented 4 years ago

@schlafly, what is the status of this PR?

schlafly commented 4 years ago

I think this PR is still correct and potentially valuable for people studying fiber offsets. This was intended to ease specsim-based fiber dither analyses, but that's all been happening in my desicmx/analysis/dither code now, so any urgency this PR had has gone away. @dkirkby, do you want this functionality, or should I drop the PR? Thanks!

desihub / specsim

Allow fiber loss calculations to be made in xy space rather than radius & azimuth space. #107