Closed hayalaso closed 3 years ago
I tested the changes using both HealpixROIMap
and HealpixConeMap
. I also used the script hal_hdf5_to_fits.py
and it seems it works for both of them. I also noticed, although probably not that important, that tables 3.5.2
or bigger was needed. Maybe not necessary in py3.
@hayalaso is this ready for review or are you still working on it?
Could you add something to the tests as well?
I got busy with other things the past couple of weeks so I don't remember. I'll need to double-check.
Btw, on github, using WIP
in the pull request name doesn't do anything. Instead, you can mark pull requests as draft: https://github.blog/2019-02-14-introducing-draft-pull-requests/ You can also convert to draft later, there's a link on the upper right, underneath where you can select reviewers.
I added a test to write a model map using the HealpixROIMap and also read it back. I think it should be ready to merge.
Hi @hayalaso!
Thank you for implementing this!
Please pull in the most recent version of the hawc_hal master into your branch and resolve any conflicts. Note that the tests directory moved to hawc_hal/tests.
Once the tests pass, this is ready to merge.
Fixes #22
Fixes #55
Hm. This isn't working for some reason. Maybe the author has to do it? Might need to close those issues manually.
Moving the
roimap
array as a table instead of being in the header when usingHealpixMapROI
.The current patch can write hdf5 files now. Reading the HDF5 using the script
hal_hdf5_to_fits.py
gives a bit of problems since hal expects to read theroimap
from the header. I'll see if this can be modified somehow.