downloaded CV datasets miss their metadata and metadata schema

[tilfischer]

Dear all,

This is connected to #10 and #32.

For analysis https://dx.doi.org/10.14272/UVLHGRADYISRGZ-UHFFFAOYSA-N/CHMO0000025.18 one might select dataset ADO-1_CuTMGqu2PF6 and download selected files. The same pop up modal also displays the metadata following the generic dataset template for CV v0.4 from LabIMotion Hub.

User on https://dx.doi.org/10.14272/UVLHGRADYISRGZ-UHFFFAOYSA-N/CHMO0000025.18 may also select the green "download metadata" button and download metadata as an XLSX file. An Excel approach is always handy for human readers, but as the metadata is defined with a JSON schema, also the metadata as JSON could be provided. One option could be a drop down menu to select the format.

Next to the green button is a blue button called "download data + metadata". The data is downloaded as a ZIP. This ZIP contains data shown in the previously mentioned pop up modal, a dataset_description.txt and another ZIP but not the metadata. Having a look to the additional ZIP, this is a BagIt Bag with a folder "metadata" not holding the metadata following the CV v0.4 schema but (it seems) that the converter.json is the converter profile. I checked the UUIDs in the converter.json with the the profile available here and it has the same UUID.

I would suggest to save the CV metadata following the CV v0.4 JSON schema in the "metadata" folder as well as the JSON Scheme (from LabIMotionHub).

Best,
Tillmann

PS: The BagIt implementation might need an update. I will open another issue on that and link accordingly.