Paton, N.W. and Gray, P.M.D., Optimising and Executing Daplex Queries Using Prolog, The Computer Journal, Vol 33, No 6, 547-555, 1990.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....different dialects of comprehensions have found their way into various programming and query languages. For example, they have been applied to imperative languages such as Napier88 [27] Not surprisingly, comprehensions are popular among systems that adopt the functional data model, such as [30,33]. The algebra underlying comprehensions has been the starting point for the theoretical studies of query languages [7,17] Its role as an internal representation language for the ODMG 93 OQL [9] is investigated in [19] Object comprehensions differ from other dialects of comprehensions in that ....

.... as an internal representation language for the ODMG 93 OQL [9] is investigated in [19] Object comprehensions differ from other dialects of comprehensions in that they are designed to be a user query language for object oriented databases [14] as opposed to just a vehicle for optimisation as in [30,19] or a notation for studying the theoretical foundation of query languages as in [7,17] The design criteria adopted are thoroughly discussed in [15] A comparison of object comprehensions with ONTOS SQL [29] OSQL [24] O 2 SQL [3] ORION [21] EXCESS [8] OQL[C ] 6] CQL [16] and XSQL [20] ....

N.W. Paton and P.M.D. Gray. Optimising and Executing DAPLEX Queries using Prolog. The Computer Journal, 33(6):547--555, 1990.

....are based, becomes clear when side effecting qualifiers are taken into consideration. It is difficult to see how SQL based languages can be extended in a similar way to cope with side effects. List comprehensions are also included in a new functional database language called PFL [SP91] In P FDM [PG90] DAPLEX queries are translated to an abstract form of list comprehensions with which optimisation is carried out. The authors commented that list comprehensions allowed queries to be expressed declaratively while DAPLEX had a navigational style of querying. It is interesting that the ....

N.W. Paton and P.M.D. Gray. Optimising and Executing DAPLEX Queries using Prolog. The Computer Journal, 33(6):547--555, 1990.

....t is the type of its head expression. Developing optimisation techniques for comprehensions is important because comprehensions provide a unifying query formalism for functional, relational and object oriented languages. Wadler [18] defines many equivalences for comprehensions. Paton and Gray [10] translate DAPLEX queries into set comprehensions and discuss their optimisation. Trinder [16, 17] gives a translation of the relational calculus into list comprehensions. Buneman et al. 4] discuss expressing SQL in the comprehension language CL and review previous work on the expressiveness of ....

N.W.Paton and P.M.D.Gray, Optimising and executing DAPLEX queries using Prolog, The Computer Journal, 33(6), pp 547-555, 1990.

....For example, the head of a set abstraction corresponds to the SELECT clause of an SQL query, the generators are correspond to the FROM clause, and the filters to the WHERE clause. Also, Trinder [Tri89] gives a translation of the relational calculus into list (as opposed to set) abstractions, [Pat90] notes that DAPLEX queries are easily translated into set abstractions, and in previous papers e.g. Pou93] we have observed the syntactic and semantic correspondence between set valued functions such as father and anc and the analogous Datalog predicates. However, set abstractions are just ....

....some provide only limited facilities for optimising user defined data types; while others allow few algebraic transformations to be applied to an expression without changing its value. Finally, the optimisation of functional database languages has been examined by several other researchers e.g. [Tri89, Bee90, Erw91, Pat90, Hey91]. Trinder [Tri89] advocates analogues of abs 1 3 for list (as opposed to set) abstractions. These equivalences are justified by assuming bag equality over lists. Implicit assumptions made are that lists are finite (otherwise, the equivalent of abs 1 would not hold for example) and that functions ....

[Article contains additional citation context not shown here]

Paton, N.W. and Gray, P.M.D. Optimising and executing DAPLEX queries using Prolog, The Computer Journal, 33(6), 1990.

....that are required. Translation into Query Internal Form (QIF) GRAIL queries are intrinsically nested structures. However, nested language structures generally imply some evaluation order, so we follow a number of earlier query planners in unnesting the source query prior to query optimisation (Paton 1990, Fegaras 1997) The QIF is a list of query components, each of which is a tuple (Base, Variable, Criteria, Cost, Cardinality) representing the evaluation of part of the query. Base is the base concept of the component, Variable is the name of the variable used to store values retrieved as a ....

Paton, N.W. and Gray, P.M.D., Optimising and Executing Daplex Queries Using Prolog, The Computer Journal, Vol 33, No 6, 547-555, 1990.

....Its output is a parse tree, constructed of nested Prolog terms, which describes the entire input program. This is then passed to the ICode generator, which transforms the full parse tree into the simpler intermediate code format. This format, which we call ICode, is based upon ZF notation [26, 18]. It retains the semantic content of the parse tree representation, while discarding much of the accounting and type information required only for parsing. It describes only the essential semantics of the input program, while the parse tree also contains information about exactly which syntactic ....

....semantics between modules, and as a storable but declarative representation for making behavioural information (i.e. methods) persist. 4 The Intermediate Code Format The intermediate code format used in the current Daplex compiler is an extension of that designed for the original compiler [26], which was based on ZF notation 1 . In this notation, a set is described by a pattern, giving the form of the elements of the set, and a declarative speci cation of what values are set members. For example, the set of third year students is described by the ZF expression: s j s student; ....

N.W. Paton and P.M.D. Gray. Optimising and Executing Daplex Queries Using Prolog. The Computer Journal, 33:547-555, 1990.

....The constraint requires every structural segment which is not preceded by another segment to be the first structural segment of its chain. Although our compiler transforms Daplex programs into Prolog procedures, Daplex itself is a functional language, with a semantics based on ZF expressions (Paton 1990). ZF expressions are a form of syntactic sugar for the # calculus (Landin 1966) which give a concise way to represent sets. For example, the following ZF expression describes the set of large proteins, for which we gave the Daplex description earlier: p p protein; molecular weight(p) ....

Paton, N.W. and Gray, P.M.D. (1990) Optimising and Executing Daplex Queries Using Prolog. The Computer Journal, 33, 547--555.

....Its output is a parse tree, constructed of nested Prolog terms, which describes the entire input program. This is then passed to the ICode generator, which transforms the full parse tree into the simpler intermediate code format. This format, which we call ICode, is based upon ZF notation [21, 15]. It retains the semantic content of the parse tree representation, while discarding much of the accounting and type information required only for parsing. It describes only the essential semantics of the input program, while the parse tree also contains information about exactly which syntactic ....

....semantics between modules, and as a storable but declarative representation for making behavioural information (i.e. methods) persist. 4 The Intermediate Code Format The intermediate code format used in the current Daplex compiler is an extension of that designed for the original compiler [21], which was based on ZF notation 2 . In this notation, a set is described by a pattern, giving the form of the elements of the set, and 2 This terminology is taken from [25] Subsequently, the terms list comprehension and set comprehension have been applied to the same notation. a ....

N.W. Paton and P.M.D. Gray. Optimising and Executing Daplex Queries Using Prolog. The Computer Journal, 33:547--555, 1990.

....The constraint requires every structural segment which is not preceded by another segment to be the first structural segment of its chain. Although our compiler transforms Daplex programs into Prolog procedures, Daplex itself is a functional language, with a semantics based on ZF expressions [22]. ZF expressions are a form of syntactic sugar for the calculus [19] which give a concise way to represent sets. For example, the following ZF expression describes the set of large proteins, for which we gave the Daplex description earlier: p j p protein; molecular weight(p) 2000] The ....

N.W. Paton and P.M.D. Gray. Optimising and Executing Daplex Queries Using Prolog. The Computer Journal, 33:547--555, 1990.

....Algorithm) comparing it with TREAT. 4. 1 Implementing Production Rules using ECA Rules The strategy presented here is based upon a prototype implementation in the OODB ADAM [31] and utilises two existing components of ADAM a query optimiser, which is a modified version of that presented in [33, 26], and an active rule system which supports ECA Rules [17] The algorithm for generating ECA rules from production rules is given in figure 5. The algorithm, given a production rule R, generates a set of ECA rules which implement R. The following are used in the algorithm: ffl The functions ....

N.W. Paton and P.M.D. Gray. Optimising and Executing Daplex Queries Using Prolog. The Computer Journal, 33(6):547--555, 1990.

....an update command into the declarative, side effect free loops part of an Daplex program. In fact, this is not the case, and the change in the semantics is less significant than might at first be thought. The semantics of a deterministic Daplex program are equivalent to a ZF expression [22, 12], which describes a set of tuples of values, where each tuple represents one of the combination of values produced by the loops 2 . For example, the loops part of the example query given on page 3 are equivalent to: c, q) j c [1. 8] q queen; column(q) c ] The generators and ....

N.W. Paton and P.M.D. Gray. Optimising and Executing Daplex Queries Using Prolog. The Computer Journal, 33:547--555, 1990.

No context found.

Paton, N.W. and Gray, P.M.D. Optimising and executing DAPLEX queries using Prolog, The Computer Journal, 33(6), 1990, pp 547-555.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC