Closed christian-oreilly closed 6 years ago
Notes:
Only NeuroCurator v0.4.4 crashes. Previous versions do not. This is because PyQt5, introduced in NeuroCurator v0.4.4, handles differently Python exceptions compared to PySide (used in previous versions of NeuroCurator).
The Zotero field 'name' is not considered the same way as 'lastName' in ZoteroWrap because it also contains 'firstName'. Also, having 'lastName' and 'firstName' filled instead of 'name' is what is expected.
When the field 'name' is used in Zotero, users should modify the reference in Zotero. It is not the purpose of NeuroCurator to add logic to handle this. Even if it was, it would introduce inconsistencies: last names are expected but 'name' contains other parts of the name.
When the new ZoteroWrap has been designed, no Zotero reference was using the field 'name'. As of 20.08.18, only one does (DOI: 10.1113/JP274049) among 751.
To see what is the behaviour of the impacted methods of ZoteroWrap when the field 'name' is used in Zotero instead of 'lastName' (and 'firstName'), please refer to the updated tests by searching for tests with ID 'no_last_name'.
NAT version: 0.4.2
Unexpected behaviour: Crashes when the interface of Neurocurator tries to display a record where the author name is not decomposed in name and surname.
Code to reproduce the behaviour:
Stack trace: