ProjectNext.items (of type
ProjectNextItemCollection is pageable, but when looking up specific items’ IDs it can be a lot of data to download and filter client-side. It would be nice to be able to filter by item state or even custom fields. I see a lot of projects - including GitHub CLI’s that don’t tend to prune items in a “done” state so that list just continues to grow. Being able to filter out closed items - since both
PULL_REQUEST items have a
state field - would help reduce bandwidth requirements.
afaict, pretty much none of the github apis (especially the graphql ones) are designed to be client/bandwidth friendly.
There are dozens of APIs where I’d love to be able to do things like this. But, it seems like a design decision – my guess is that it’s cheaper for github to stream data out than to spend cpu cycles performing filtering and worrying about side-effects from that filtering.