This is a bug I just discovered with the way MakeCorr2FileKwargs works. At the moment, if it has to write some files to disk, it only writes the fields that are in all the data sets it's writing--but that's a silly thing to do when performing, eg, a galaxy-galaxy lensing computation, where we only have positions and not shears for one of the two data sets. So right now any point-shear correlation functions fail, because it doesn't write out the shears for the second data set. Whoops!
This is a bug I just discovered with the way
MakeCorr2FileKwargs
works. At the moment, if it has to write some files to disk, it only writes the fields that are in all the data sets it's writing--but that's a silly thing to do when performing, eg, a galaxy-galaxy lensing computation, where we only have positions and not shears for one of the two data sets. So right now any point-shear correlation functions fail, because it doesn't write out the shears for the second data set. Whoops!I'll fix this on this branch.