Closed mkelley closed 5 years ago
Thank you for the report and especially for the standalone example. 👍 Very easy to verify and it made it pretty easy to debug.
Actually there are two issues here: One is that you are correct and the ccds
-generator doesn't update the original. However the second problem is that ccd += 1
creates a numpy-array. So you didn't modify ccd
but just created a new object.
I'll submit a fix for the first issue but the second one is impossible to fix without creating in-place arithmetic operations (like __iadd__
).
OK, now I get it, ccd.data += 1
would be the right call.
@mkelley -- is it ok to close this now?
Closing this, but please re-open or create a new issue if something more is needed.
Hello,
When using the
CCDData
generator and theoverwrite=True
option, changes to the returned object are not updated in the FITS file. However, the hdu generator does work as expected:Results:
The saved CCDData mean is 0.0, but I expected 1.0.
astropy 2.0.dev17751, ccdproc 1.3.dev1066, and numpy 1.11.0.
Thanks for the help, Mike