Type-theoretic Means for Querying Heterogeneous Data Sources Available via Cloud APIs
A.I. Domracheva, Pavel Shapkin · Procedia Computer Science · 2016
We propose a unified approach for querying data sources which are accessible in form of heterogeneous APIs rather than some standards compliant databases. Different APIs use different query structures and support different capabilities. Type-theoretic approach is used for query structure representation. Because of the fact that not any API will support the full expressibility of the query language we propose techniques for “widening” the query to make it compatible with the given API possibilities. After the query is executed additional filters are applied to the acquired data. In order to achieve this goal a special set of query expansion rules was developed. It consists of the rules that generate special conversion functions which transform more specific queries into less specific ones.