Closed mdoering closed 5 months ago
Should https://api.dev.checklistbank.org/dataset/9847/sector/publisher
list publishers for a project then?
Yes. For a project or release. Management should only be possible for the project and release dataset keys give you access to the archived version as it was at release time. Im still working on that though.
Sorry, fixed the dev db which wasnt updated. works now but is all empty yet
What does the datasetKey
refer to?
datasetKey is the project or release key the sector publisher belongs to. This is not a global publisher list, just the set of publishers that are configured in a project to create sectors automatically and act as an aggregation source. Hence this can change from project to project and release to release
I have created the publishers for COL on dev as they should be at this stage: https://api.dev.checklistbank.org/dataset/3/sector/publisher
datasetKey is the project or release key the sector publisher belongs to.
Sure. It is the key from the path parameter obviously
For projects we need to manage sector publishers, i.e. publishers of datasets that sectors are automatically created for in xreleases and which we use to aggregated sectors & metrics as "sources".
I would propose a new "publisher" tab in Options. As there will only be a handful publishers we should be able to use a table similar to users. Adding a new publisher should query the GBIF registry and select a publishers UUID and title from it. The title and alias (required) can be changed by a user, only the uuid key is fixed.
The CRUD API is at
/dataset/{key}/sector/publisher
with the following Publisher class:for each publishers there should be a link to a) the GBIF publisher page and b) the dataset search with that UUID as a filter.
It would also be good to show 2 counts:
withoutSectorInProject={projectDatasetKey}
and allows to preview the new datasets before a release is done.