If I understand correctly, you get the same error with both ntuple->PrintInfo(); or ntuple->LoadEntry(0);, is it correct? Could you please provide an example ntuple to reproduce your error?
Yes I get the same errors with both ntuple->PrintInfo(); and ntuple->LoadEntry(0);
I have tried with multiple files with RNTuples and I seem to get the same errors (although when I create my own RNTuple and read from that I do not seem to get any errors)
One of the files I tried can be accessed here, where I read the EventData RNTuple: https://we.tl/t-RvDoTHALfB
In my case, only ntuple->LoadEntry(0) raises the exception, which looks like an IO error, while ntuple->PrintInfo() works smoothly. Let me add in the loop @pcanal.
I did not create the RNTuple, but I got it from here: https://atlas-art-data.web.cern.ch/atlas-art-data/grid-output/main--dev3LCG/Athena/x86_64-el9-gcc13-opt/
I have also later tried with ROOT version 6.34, with this asetup: asetup Athena,main,latest,gcc13
The one I used before, where I got the errors, was this (where the ROOT version is 6.35.01): asetup Athena,main--dev3LCG,latest,gcc13
It looks like your issue might be related to improper typename normalization by the RNTuple internals, for which a fix was merged recently (see [ntuple] fix type name normalization by jblomer · Pull Request #17637 · root-project/root · GitHub). The RNTuple you sent us was created prior to this merge. Could you try with a file that was produced using a more recent version of the ROOT master branch (after Feb 14, to be specific)?
Thanks for checking, I indeed still observe the error as well. This will require some more investigation on our side, I will let you know once we understand the issue better.
The issue you reported is related to the way RNTuple internally resolves type names, and a fix has been provided by @hahnjo. The reproducer you provided now works for me without errors, using the latest ROOT nightly (i.e., using asetup Athena,main--dev3LCG,latest,gcc13). Could you check if for you it now also works as expected?