Closed astrochun closed 2 years ago
This is not a bug, this is at most unclear documentation. The documentation for AuthorRetrieval().get_documents()
states:
Parameters args (str) – Parameters to be passed on to ScopusSearch().
kwds (str) – Parameters to be passed on to ScopusSearch().
Why is that not obvious?
Hi @Michael-E-Rose, thanks for the response. The reason why it wasn't obvious is because I had thought refresh
from the AuthorRetrieval
was an instance variable, such that if I say refresh=3
, any get methods would have those settings by default.
I suppose this is more of a feature than a bug for the default settings in the get methods to be consistent with input arguments when the object was created.
What about we add a note Note: The class' `refresh` parameter is not used here.
?
What about we add a note
Note: The class' `refresh` parameter is not used here.
?
That would suffice.
pybliometrics version: 3.3.0
This problem occurs when attempting to update the Author record based on an
AuthorRetrieval
withrefresh=True
. The use of the.get_documents()
does not trigger a newScopusSearch
. Thus, records can be outdated.Code to reproduce the bug:
The difference is due to one new paper that appeared after the first caching. Currently, the API requires providing
refresh=True
, which is not so obvious - took me a while to realize that the document search query did not update accordingly.Expected behavior: If certain arguments are set at the class level, they should be reflected in method calls to ensure that everything is in sync.