Good idea that also holds generic across other ID columns (eg. Subject: Re: Linking locationIDs to existing ( #5) It should not be necessary to have a “human readable locationID”, the uuid “locationID” is already well documented by the content of the other columns, the xy coordinate in association with locality, verbatimLocality, stationNumber and siteNumber as well as the other columns are already a secondary key/id to the locality geographical point, I am afraid that generating another key as serial human number starting from 1, 2, 3… to keep track of locality names will overload register with unreadable number when having several thousands localities.įrom: Anders G. My suggestion is that we start a new repository on the githubsite /NTNU-VM called NaTron and start issue tracking of db specific generic changes there. We are likely to run into several other issues in time when starting to using the NaTron database for real. verbatimLocation is a candidate field (I much prefer to recirculate DwC terms than to invent new ones if possible). This must then have a unique constraint across the location table.

For datamanagment/natron structure in general: There should be identified a field that should be used for the human readable locationID.For natronbatchupload: Probably best to keep these functions separated in own compartment (mapping-part).If adding location then the UUID for the location (locationID) should be set in the original data-set and one first have to upload the locations etc. If not, one have to add the location first. However, this means that the location are registred in the database up-front.

Ffxiv natron location