Problem in published Dataset related to tabulartree

  1. The dataset https://nomad-lab.eu/prod/v1/staging/gui/dataset/id/UfbrNqwOQrOPs3SURIIhqw have some entries whose data.archive.json files contain illegal NaN values set for some fields. This causes to get an error in the frontend not a valid JSON

It is probably part of a bigger problem, as in python for example, one can export a json formatted string with this illegal NaN values in this way:

test = {"test": float('nan')}
json.dumps(test)

but this is not a valid format of a JSON string. All the entries in the upload wZzlkn24S-a8O1s33owOOg have this problem. After replacing the NaN values in the entries with null, a reprocessing is needed.

  1. The upload with upload_id wZzlkn24S-a8O1s33owOOg has a problematic schema in which there exists duplicate quantities under the same section (for example surface_content_of_O_(atom_%)_O quantity)

  2. There are 4 schemas in this dataset (XRF, XRD, TPRO and BET schemas) but None of them are populating tabulartree

  3. Some other entries (for example from upload_id 'qegdLWpzQ7yAFmYikFNydg') complains that HistoryCard cannot be rendered which is probably because the processed data is old and they don't contain section_defs in their metainfo. Reprocessing the entries should fix this. But in general, it is better to correct line 73 of the HistoryCard from !index.section_defs.some to !index.section_defs?.some.

Edited by Amir Golparvar