Electrodes table "group" display changed?

[weiglszonja]

In the electrodes table "group" is showing up again like this:

While it should be the link/path to the ElectrodeGroup this electrode belongs to.

https://neurosift.app/?p=/nwb&url=https://api.dandiarchive.org/api/assets/8a2039bb-d84d-4a35-a897-d9d630dbf61d/download/&dandisetId=000947&dandisetVersion=draft&tab=neurodata-item:/general/extracellular_ephys/electrodes|DynamicTable

[magland]

@weiglszonja This is now partially fixed. At least the number of rows is correct:

However, the references in the group column are just arrays of size 8 (representing 64-bit values). There's no way in the browser to resolve which objects these are referring to.

As a side note, when the dataset has been prepared with lindi as in this example: https://neurosift.app/?p=/nwb&dandisetId=000947&dandisetVersion=draft&tab=neurodata-item:/general/extracellular_ephys/electrodes%7CDynamicTable&url=https://api.dandiarchive.org/api/assets/766438c5-fa8c-42a5-bd26-41c8a89be93a/download/

The references do show up.

[weiglszonja]

thank you @magland for taking a look and for the explanation. I'm not familiar with LINDI yet, so the reason the first snippet that I sent doesn't show the references is because it had not been prepared yet?

[magland]

thank you @magland for taking a look and for the explanation. I'm not familiar with LINDI yet, so the reason the first snippet that I sent doesn't show the references is because it had not been prepared yet?

Yeah. I have a gh action that iterates through all the dandi nwb files and prepares LINDI files (json/zarr files) in stores them in a cloud bucket. When neurosift points to one of these files on dandi, it first checks the LINDI bucket. The example you shared has not yet been indexed.

You can read more about LINDI here.