Closed arthur-shaw closed 2 years ago
I second this enhancement. The current function is quite heavy I feel.
In addition to being able to return only a selection of nodes, it would also be very useful to make use of the where: InterviewsFilter
, e.g. to only get interviews from questionnaire X. In fact, I'll need this for an upcoming project, so I'll try to fork next week and give it a try.
P.s. I also often stumble over the issue that the function returns the following
Error in get_interviews_by_chunk(workspace = workspace, take_n = chunk_size, : Object reference not set to an instance of an object.
which might be due to the server not being able to handle the IOPS (?) . I'll try to find the time to investigate further and maybe raise another issue =)
Thanks, @petbrueck.
I also often get the same error as you describe above. If you have time to investigate the issue, I'd love to understand better what's going on. For a current project, I end up having to rerun get_interviews()
several times before it works. This is both frustrating and strange. Frustrating because I have to run the script repeatedly and outside of a script for things to work. Strange because the function should, in principle, either work or not work systematically--not work sometimes but not others, for the same scenario.
I have rewritten get_interviews_by_chunks()
using GraphQL. I just need more testing before releasing. Perhaps I can do a pre-release as a branch. Stay tuned.
The
get_interviews()
function fetches all available data for an interview. There needs to be a function to fetch user-specified attributes about interviews. The GraphQL offers the following data. The user should be able to request particular attributes rather than fetch them all and select those that are relevant. This may be a new function or a change in the interface, where the full payload is the default but fewer attributes can be requested.