References in periodicals archive ?
1987], and tools like flex and bison can be used for the construction of SQL or OQL parsers just as for most other programming languages.
A client ships the SQL (or OQL) code of a query to the server; the server evaluates the query and ships the results back to the client.
As we will demonstrate in this article, our calculus captures most features of OQL and is a good basis for expressing various optimization algorithms concisely.
Section 3 gives the translation of some OQL query forms into monoid comprehension.
Several recent proposals for object-oriented database languages, including OQL, support multiple collection types, such as sets, bags, lists, and arrays.
When used in OQL queries, a binary relationship between the classes A and B offers two ways of relating A and B objects: given an instance of A to retrieve the related instance(s) of B and given an instance of B to retrieve the related instance(s) of A.
Nearly all OQL expressions have a direct translation into the monoid calculus, with the exception of indexed OQL collections.
In this paper we consider the most important form of OQL queries: select expressions.
A surprising and really disturbing result is that OQL queries cannot be type-checked in the type system of the Java binding of the ODMG standard either.
This leads to a loss of type information and makes type-checking OQL queries impossible.
We prove that if explicit type casts of control variables down the inheritance relationships are required in OQL queries, type-checking OQL queries becomes possible.
In contradistinction to the situation described above, a positive result that we established formally is that type-checking OQL queries presents no problem for a type system with the basic form of parametric polymorphism (universal type quantification).
Acronyms browser ?
Full browser ?