Some datasets has combination of numeric and decimal values which will make it harder to query OR one has to cast the datatype for observation with different datatypes before making query. In order to avoid it and makes it easier to query, do the following :
Are there any non-numeric values (e.g. string) in the data ? Generate logs while the script is running to identify which datasets has non-numeric datatypes.
Would adding decimal types to all observation values (e.g. converting numeric values into decimal) will be beneficial ?
Whether all the datasets can be represented as xsd:integer or xsd:decimal.
aftiqb
commented
12 years ago
This issue is a part of https://github.com/LATC/EU-data-cloud/issues/1 :
Some datasets has combination of numeric and decimal values which will make it harder to query OR one has to cast the datatype for observation with different datatypes before making query. In order to avoid it and makes it easier to query, do the following :