Document Server@UHasselt >
Research >
Research publications >

Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/25921

Title: J-Logic: Logical Foundations for JSON Querying
Authors: Hidders, Jan
Paredaens, Jan
Van den Bussche, Jan
Issue Date: 2017
Publisher: ACM
Citation: PODS 17 Proceedings of the 36th ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems, ACM,p. 137-149
Abstract: We propose a logical framework, based on Datalog, to study the foundations of querying JSON data. The main feature of our approach, which we call J-Logic, is the emphasis on paths. Paths are sequences of keys and are used to access the tree structure of nested JSON objects. J-Logic also features “packing” as a means to generate a new key from a path or subpath. J-Logic with recursion is computationally complete, but many queries can be expressed without recursion, such as deep equality. We give a necessary condition for queries to be expressible without recursion. Most of our results focus on the deterministic nature of JSON objects as partial functions from keys to values. Predicates defined by J-Logic programs may not properly describe objects, however. Nevertheless we show that every object-toobject transformation in J-Logic can be defined using only objects in intermediate results. Moreover we show that it is decidable whether a positive, nonrecursive J-Logic program always returns an object when given objects as inputs. Regarding packing, we show that packing is unnecessary if the output does not require new keys. Finally, we show the decidability of query containment for positive, nonrecursive J-Logic programs.
URI: http://hdl.handle.net/1942/25921
Link to publication: http://alpha.uhasselt.be/~lucp1080/raleigh.pdf
DOI: 10.1145/3034786.3056106
ISBN: 9781450354752
Category: C1
Type: Proceedings Paper
Appears in Collections: Research publications

Files in This Item:

Description SizeFormat
Published version610.23 kBAdobe PDF
Proof of Peer review - Review Reports176.93 kBAdobe PDF

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.