Todo
check the module and its the documentation
Tests for tan_dir() and tan_inv() need values obtain from external codes as reference
Todo
decimal2dms(): the two external if branches are almost exact repetitions. Solution: save the sign of the input decDeg variable; convert the input as it were positive, the add the sign before returning. Also most of the internal if branches are either superfluous or repetitions that can be factored out.
Also decimal2hms() show a fair amount of repetitions, in big part similar to the above function.
Once the refactoring is finished, the tests should be updated to cover all the branches.
hms2decimal() tests fail: all the input values in the setup_class method are obtained with ned calculator. The values obtained with the function disagrees at the 7th digit level with the value from ned (should we consider it a failure?). Using only the hour, the values is wrong.
Should we use astropy.coordinates instead of this implementation?
Todo
angular_separation_deg(): the test fails. Where it is possible to get a reference value? http://cads.iiap.res.in/tools/angularSeparation and http://www.asdc.asi.it/dist.html gives very different answers, and none of them agrees with the output of the function
Todo
a module in pyhetdex should not depend on the structure of a class defined in code that uses pyhetdex. Here referring to shot
Todo
a module in pyhetdex should not depend on the structure of a class defined in code that uses pyhetdex. Here referring to shot
either pass the illumination_model option as parameters to fplaneToThroughput() or check that it is implemented when initialising ThroughputServer.
Todo
Improve the documentation.
ifu_centers: plotFocalPlaneQuicklook() and get_image() have a ifu_centers arguments, which is the center of the IFUs in the focal plane. This is confusing as in HETDEX jargon IFUCen refers to the fibers within one IFU.
Write tests.
Use io.StringIO instead of StringIO.StringIO in python2
The CD matrix could be substituted by astropy WCS:
Check if io.StringIO works for python2, if yes, we don’t need it pyhetdex.tools.six
Todo
filename: is hard coded. Instead of the output directory pass the full output name in the function arguments
Rename the function: probably plot_image would be more appropriate
box, size_cut_out: they should not be hard coded
html size should not be hard code
Todo
ifu_size is hard coded to 0.012. Move to optional argument
Rename the function: function does not plot!
Todo
either catch errors from urlopen() in retrieve_image_SDSS() and retrieve_image_DSS() or add the notion that it’s raised and do something about it in get_image()
Todo
use Yields instead of Returns when the numpydoc 0.6 update will be available
Todo
Warn the user when getting an existing name with the other options not set to the defaults
Add context manager?