Hey folks - I’m working on generating some wide names for the current structure for how the European Wastewater Observatory will ingest wastewater surveillance data from member states. Shout out to @Sorin for making this happen!
For the most part, the wide names are easily made for all fields. However there are 4 headers that present a challenge:
instrument_name_flowmeterinstrument_name_samplerorganization_name_Laborganization_name_WWagency
All of these would ordinarily fit under the standard “exceptional attribute” type of wide name, following the structure of tableShortName_n_aggregation_attribute1_..._attributeN. However in all four of these cases, the second attribute here is already populated. Essentially, instrument_name_flowmeter is recording the in_name and in_insType, but the insType value is already specified as either a “sample” or “flowmeter”.
This is hardly what I would consider an entirely shocking format, but it does mean we need to come up with an alternate structure.
The considerations here are that ideally this would stay somewhat true to the original structure, and also be potentially machine readable. Already with the integer and the aggregation (‘AND’ or ‘OR’) we have something pretty strong. I think perhaps what we need is a similar tag in the header to advertise that one is populated. For example:
instrument_name_sampler → in_2_AND_name_insType_$sampler$ or in_2_AND_name_insType__sampler_
where the $ or extra _ indicate that a entry is actually populating the preceding attribute? Thoughts? @dmanuel @jeandavidt