My results are stored as TGraphAsymmErrors in TCanvas'es saved as C macros. I noticed that CFileReader does not support TGraphAsymmErrors though. Not a big issue (as a workaround, I simply dumped my graphs into txt files, then read them using numpy and used hepdata_lib on that), but I thought I would mention it, in case someone wants to add this feature for some future release of hepdata_lib, and make it cover even more use cases.
My results are stored as TGraphAsymmErrors in TCanvas'es saved as C macros. I noticed that CFileReader does not support TGraphAsymmErrors though. Not a big issue (as a workaround, I simply dumped my graphs into txt files, then read them using numpy and used hepdata_lib on that), but I thought I would mention it, in case someone wants to add this feature for some future release of hepdata_lib, and make it cover even more use cases.