Purely Relational FLWORs.
Torsten Grust
Abstract
Torsten Grust
Abstract
We report on a compilation procedure that derives relational algebra plans from arbitrarily nested XQuery FLWOR blocks. While recent research was able to develop relational encodings of trees which may turn RDBMSs into highly efficient XPath and XML Schema processors, here we describe relational encodings of nested iteration, variables, and the item sequences to which variables are bound. The developed techniques are purely relational in more than one sense: (a) we rely on a standard (or rather: classical) algebra that is readily supported by relational engines, and (b) we use relational concepts like functional and multivalued dependencies to significantly simplify the emitted plans. This work blends well with the mentioned tree encodings and thus contributes a further important building block to investigations into XQuery processors based on relational database technology.
OpenAlex reports 27 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
We report on a compilation procedure that derives relational algebra plans from arbitrarily nested XQuery FLWOR blocks. While recent research was able to develop relational encodings of trees which may turn RDBMSs into highly efficient XPath and XML Schema processors, here we describe relational encodings of nested iteration, variables, and the item sequences to which variables are bound. The developed techniques are purely relational in more than one sense: (a) we rely on a standard (or rather: classical) algebra that is readily supported by relational engines, and (b) we use relational concepts like functional and multivalued dependencies to significantly simplify the emitted plans. This work blends well with the mentioned tree encodings and thus contributes a further important building block to investigations into XQuery processors based on relational database technology.
Key concepts: XQuery, Relational algebra, Computer science, Relational database, Codd's theorem, XPath, Conjunctive query, Relational calculus