A Practitioner’s Approach to Normalizing XQuery Expressions

Revisiting Source-Level XQuery Normalization

“…XML has emerged as a standard for representing, storing, and exchanging data on the Internet, and XPath and XQuery are widely used for querying XML data [8], [12]- [14], [24]. XQuery encompasses XPath and supports a FLWOR expression that is similar to a select-from-where expression in SQL.…”

“…In his seminal paper, Kim [11] opened the area of unnesting nested queries in the relational context. With the advent of the XQuery language, several normalization rules [2], [4], [12], [15], [16] have also been reported in the XML context.…”

“…Examples of the former include the rules proposed by Kim [11], Manolescu et al [15], Lee et al [12], and Brantner et al [2], and examples of the latter include those proposed by Fegaras et al [5] and May et al [16].…”

“…Second, there is no uniformly accepted algebra for querying XML data yet [2]. Third, the algebras proposed in the literature [5], [16] are too complicated to implement practically in conventional DBMSs [12]. Nevertheless, existing source-level rules have the following problems.…”

“…First, they do not handle quantified expressions. Second, they incur duplicated query results [1], [12]. Third, unnested queries obtained by the source-level normalization rules use many temporary files to decouple the execution of the inner query from that of the outer query [1], [3], [7], [10]- [12], [17], [21].…”