Pierre-Alexandre Fonta

What was the reference ID? I wasn't able to reproduce this behaviour with NeuroCurator v0.4.0 (packaged) and the following steps: 1. In tab _References_, double-click on the reference of ID...

The issue appears when _Relationship_ in _Parameter details_ has a value different from 'unspecified' ('point', 'directed', 'undirected'). When creating a new parameter, it is possible that _Relationship_ has already a...

There is likely the same bug when we add new values instead of modifying existing ones.

Attempting to create a new annotation of type 'position' gives the following stack trace. It might reveal the origin of the bug: Ghostscript is not found. > wand.exceptions.DelegateError: FailedToExecuteCommand `'gs'...

Hello @MFSY. The template shouldn't have a nested value for `type`. It seems forge.template(...) for RdfModel and with the example configuration (Neuroshapes) doesn't work. The output seems to be always...

Hello @MFSY After some investigations, the `type.id` is required now for top level `type` but not for `type` of nested entities. This `type.id` introduces then an inconsistency. So it seems...

By the way, `test_create_templates()` is part of the tests which are failing for `RdfModel`. This support the idea that something went wrong for `template(...)` for `RdfModel`.

Following the discussion: For when the path is known, would the following UX suit your use case? ``` f = forge.traverse(a, 'b.c.d.e.f') ```

There is actually another possibility: using the `forge.paths(...)` returned object. The great thing is that users would have autocompletion and it wouldn't be a string. However, there are some conditions....

Hello @jdcourcol and @joni-herttuainen, Following our discussion I have tried to synthesize the design for a prototype. Please let me know if I have captured them incorrectly or if you...