METHOD AND APPARATUS FOR PROCESSING MARKUP LANGUAGE SPECIFICATIONS FOR DATA AND METADATA USED INSIDE MULTIPLE RELATED INTERNET DOCUMENTS TO NAV

I claim:

1. In a network of computers over the internet, a multi-tier client/server system comprising of:

A) Clients with browsers capable of processing documents in Extensible Markup Language (XML) and Resource Description Framework (RDF) carrying data/metadata information for electronic commerce and electronic business where these clients make synchronous communications through Hypertext Transfer Protocol (HTTP) with various internet sites and with object relational database servers at multiple locations consisting of XML/RDF integration software and software compliant with Common Object Request Broker Architecture;

B) Clients with browsers capable of creating and maintaining thin client windows on demand for persistent connectivity through Internet Inter ORB Protocol (IIOP) with remote object relational databases over the web for performing transactions;

C) Relational database servers with schema and business application logic defined in the form of object packages comprising:

a) User-defined packages for providing call specifications for a set of interfaces to embed in Structured Query Language (SQL) queries where these interfaces are implemented as methods in Java classes and are applied to records resulting from execution of relational operations;

b) User-defined packages for specifying operations over attribute values from multiple tables where interface definitions for methods and operators carry parameter definitions along with parameter types defined directly as table names and attribute names for safe execution semantics;

c) User-defined packages specifying interfaces where a parameter type definition maps to another interface existing at a local or remote schema object on the internet;

d) User-defined packages specifying interfaces where parameter type definitions map to tables and attributes existing at a local or remote schema object on the internet;

e) User-defined packages where Uniform Resource Identifiers (URIs) are used to locate elements in remote schema objects on the web and such URIs defining schema objects are stored in data dictionaries or repositories;

f) Method and operator interfaces defined in user-defined packages implemented in Java classes;

D) Executions of application logic in object packages on local or remote schema objects performed by Object Request Broker services in Common Object Request Broker Architecture (CORBA).

2. A system as defined in claim 1 wherein said clients and said object relational database servers communicate through documents in Extensible Markup Language and Resource Description Framework as well as thin client persistent windows for transactions and wherein said multiple relational database servers further communicate among themselves through documents in Extensible Markup Language and Resource Description Framework as well as through persistent connectivity in Internet Inter ORB Protocol (IIOP), switching roles as clients and servers.

3. In a network of computers over the internet, a multi-tier client/server system comprising of:

(a) Multiple related internet sources representing documents in Extensible Markup Language (XML) and Resource Description Framework (RDF) for describing data and metadata for various resources and their interrelationships necessary for electronic commerce and electronic business;

(b) Resource Description Framework (RDF) describing properties and values of properties for resources identified by Uniform Resource Identifiers (URIs) where these URIs are used as primary keys and foreign keys in a normalized relational model of description for these said relationships over resources;

(c) Resource Description Framework (RDF) describing properties and values of properties for resources mapped to tables with foreign key/primary key relationships in one or more relational databases where foreign key attributes may be named to express same or inverse meaning of corresponding relationship names in RDF description, these names being defined in XML namespaces and schema definitions;

(d) Resource Description Framework (RDF) describing properties and values of properties for resources further comprises:

(1) RDF container types, namely Bag, Sequence and Alternative to represent collection of resources where these container types are mapped to normalized relations in relational model;

(2) RDF Bag container type with one-to-many relationships mapped to two tables in first normal form where one table contains primary key as the URI of the resource under description and the other table contains primary keys as the URIs of the resources in Bag container along with foreign key inherited from first relation;

(3) RDF Bag container type with many-to-many relationships mapped to three tables in second normal form where first table contains primary key as the URI of a resource under description, second table contains primary keys as the URIs of resources in Bag container and third table maps primary keys of first table to primary keys of second table capturing many to many relationships;

(4) RDF Bag container type with more complex nesting of relationships normalized into relations in higher normal forms;

(5) RDF collection type Alternative (Alt) mapped to two relations in one-to-one relationship where the first relation contains records with a two part primary key, first part as the URI of the resource under description and second part a tag to identify a specific alternative, and the second relation contains primary key as URI of a resource in Alternative container along with foreign key being the two part primary key inherited from first relation;

(6) RDF collection type Sequence (Seq) mapped to a single relation with multi-part primary key where first part is the URI of the resource under description and second part is URI of a resource in Sequence container maintaining sequence by sorting over compound primary key;

(e) Data and metadata information stored in a relational database or part of a relational database or union over multiple relational databases mapped to documents in Extensible Markup Language and Resource Description Framework for entities, properties and values of properties where such entities and properties are URIs identifying table names and attribute names respectively.

4. A system as defined in claim 3 wherein said documents in Extensible Markup Language and Resource Description Framework further represent queries in Structured Query Language (SQL) by defining operators and key words of SQL language in Extensible Markup Language namespace schema definitions and wherein operands to such relational operators are either resources represented by Uniform Resource Identifiers or literals.

5. A system as defined in claim 4 wherein said Structured Query Language (SQL) queries embed method interfaces represented in Extensible Markup Language and Resource Description Framework and wherein such method interface names along with all other parameter names within method interfaces are defined in Extensible Markup Language namespace schema definitions.

6. A system as defined in claims 5 wherein said object SQL queries with or without embedded method interfaces are implicitly generated as a result of property based navigation and information requirement specification from multiple related documents in Extensible Markup Language and Resource Description Framework.

7. A system as defined in claim 6 wherein said implicitly generated client queries travel to said object relational databases for collaborative execution of said queries and wherein said object relational database servers further generate documents in Extensible Markup Language and Resource Description Framework for representing results of such said queries for sending back to said clients for further navigation.

8. A system as defined in claim 3 wherein said multiple documents in Extensible Markup Language and Resource Description Framework constitute different views over a unified Database consisting of multiple object relational databases over the web and wherein such views are updated by generating new documents to replace old ones as changes are made to said databases by inserts, deletes and updates.

9. A system as defined in claim 3 wherein said metadata information inside a unified Database over multiple disparate relational databases over the web are possible to be extracted for visualizing entities with various relationships without fetching detailed values of attributes from databases and wherein such entity-relationship diagrams represented in Extensible Markup Language and Resource Description Framework are sent to client locations for visual inspection, manipulation and focused selection of entities or attributes for further detailed visualization, thereby saving critical traversal time of large contents and data values of unnecessary entities and attributes over the internet.

10. A system as defined in claim 9 wherein said entity-relationship diagrams representing metadata information over resources and relationships are visualized through software visual tools for various manipulations and wherein such tools are used to create new resources with relationships to existing entities or collection of entities, forming new web pages by implicit translations of such said entity-relationship diagrams into documents in Extensible markup Language and Resource Description Framework.

11. In a network of computers over the internet, a multi-tier client/server system incorporating distributed business application logic with relational data at disparate locations comprising of:

A) Multiple documents in Extensible Markup Language and Resource Description Framework with data and metadata definitions to view, navigate and trigger implicit queries against a logical schema definition for a single unified Database consisting of multiple disparate relational database instances with tables, foreign key/primary key relationships, user-defined types and business application logic in object packages;

B) Multiple documents in Extensible Markup Language and Resource Description Framework with data and metadata definitions for navigating through a single unified schema containing many instances of component physical schema of relational databases, so as to locate and focus on a data set(s), a text item or any other content(s) and to initiate transaction on such focused item(s) in a separately created thin client window(s) establishing persistent connectivity with any remote physical database instance(s);

C) A single unified virtual Database for navigation, transaction processing and implicit object relational query processing over the web, defined over multiple physical component schema consisting of:

a) Uniform Resource Identifiers (URIs) used for locating remote schema objects;

b) Relations as view caches to maintain relationships over foreign keys inherited from a table in the local component schema and primary keys of a table in a remote schema located by URIs;

c) Relations as view caches to enforce referential integrity across component physical relational databases over the internet;

d) Business logic in the form of object packages implemented in Java classes to incorporate processing logic over a join or other relational operations performed in a local or remote component schema;

e) View cache tables maintaining an attribute(s) which is an user-defined type with an associated object package implemented as Java class at a remote schema location and URIs as attribute values for locating primary keys in that remote schema.

D) Property based specification and navigation through multiple documents in Extensible Markup Language and Resource Description Framework for implicitly generating object SQL queries to represent a multiple resource join in a single logical schema where such a join is performed as multiple partial joins at disparate component physical schema with communications through Internet Inter ORB Protocol and Hypertext Transfer Protocol at each schema location;

E) Property based navigation through multiple documents in Extensible Markup Language and Resource Description Framework for implicitly generating and executing object SQL queries leading to results from many levels of business logic applications on a complex join in a logical schema where such a business logic application on a complex join is executed as cooperative processing of distributed business logic on partial joins in component schema at disparate locations communicating through Object Request Brokers;

F) Object SQL query processing over a single unified database over the internet in two separate stages of processing where stage (1) of processing resolves metadata definitions from disparate schema components to decompose an object SQL query into separate components and stage (2) of processing executes such SQL components in physical databases;

G) Object SQL query executions in a single unified Database over the internet leading to collaborations in stage (1) of computations by: (a) sending proper sets of records to remote schema locations, (b) receiving resulting records from remote schema locations, (c) creating temporary tables in local database for storing record sets received and (d) triggering necessary SQL components to perform join over record sets from local and remote databases.

12. A multi-tier client/server system defined in claim 11 wherein said virtual unified Database over disparate database components comprises two distinct SQL processing for stage (1) and stage (2) of computations and wherein each physical database component over the internet contains separate software layers for collaborative object SQL processing in stage (1) of computation.

13. A system as defined in claim 11 wherein said logical schema for a unified Database partitioned into said physical schema components comprises means of compiling object SQL (Structured Query Language) queries triggered directly or expressed in Resource Description Framework documents and executing such queries against the said logical schema irrespective of locations and distribution of physical schema components by the steps including:

A) Keeping non-detailed metadata (information about relational data) for the logical schema at each component physical schema locations or at a centralized repository so that locations of other component schema are available to said stage (1) computation of SQL processing for all tables, types and packages used in a query;

B) Parsing a Structured Query Language (SQL) query at a physical component schema location by said stage (1) SQL processing layer to resolve definitions for tables, types and packages whatever is locally available and extracting portion of the query for sending it to other locations for preparation by corresponding said stage (1) SQL processing layers at disparate locations wherever the relevant metadata is available;

C) Preparing fully a SQL query at multiple sites by successively parsing with locally available metadata and initializing any object packages used at each component schema location by said stage (1) SQL processing software layers;

D) Beginning execution only after complete preparations at all involved locations;

E) Performing relational operations by said stage (2) SQL processing layers and executing application logic at each component schema location by collaborations at said stage (1) SQL processing layers through Object Request Brokers;

F) Completing any business logic execution at a location depends on successive completion of other executions at other locations when there are dependencies with nested applications and generating result sets in Extensible Markup Language and Resource Description Framework by said stage (1) SQL processing layers for transporting back to client locations.

14. A system as defined in claim 13 wherein said preparation phase and said execution phase of an SQL query further comprises means of including internet locations with centralized legacy databases and other existing data stores with middleware software for processing documents in Extensible Markup Language and Resource Description Framework and also middleware software compliant with Common Object Request Broker Architecture.

15. A system as defined in claim 14 wherein said legacy databases and said existing centralized data stores carrying no object packages for business logic and carrying necessary middleware software for processing documents in Extensible Markup Language and Resource Description Framework further comprises means of supporting SQL queries with embedded business logic against such legacy databases by defining object packages in other component schema locations.

Description Submit all comments and votes

BACKGROUND OF THE INVENTION

1. Field of the invention

This invention relates to data and metadata management over the internet for navigating, querying and manipulating any kind of information by using and executing high level specifications in Resource Description Framework and by supporting multiple object relational database resources over the web. With the advancement of World Wide Web, a large number of different types of objects (text, file, audio, video, image as well as relational data) are being created everyday. Internet can be visualized as a large single database. Querying and manipulating such a large database from many different perspectives is a nontrivial task. Additionally, transactions over the web, electronic commerce with complex buyer/seller relationships and distributed many tier application architecture are also posing demand for new technology solutions. This invention relates to those specific technology needs (a) by incorporating advanced metadata specifications in extensible markup language for implicit generation of object SQL queries in conjunction with navigational capabilities and (b) by incorporating need-based persistent connectivity through object brokers to support transactions over database web entities.

2. Description of the Prior Art

Internet is becoming an important channel for retail commerce as well as business to business transactions. The number of web buyers, sellers and transactions is growing at a rapid pace. But the potential for the internet for truly transforming commerce and business still remains to be fully realized. Electronic purchases are still largely non-automated. Software techniques are required to automate several of the most time consuming stages of web surfing and buying/selling processes. Additionally, business to business web transactions are demanding seamless query facilities over all kinds of information at the front end web portal sites as well as at the back end relational databases in a connected enterprise. Uniform querying, decision support and transactional characteristics need to be present over any kind of web data despite of the fact that data may or may not be immediately present in a single relational database. So far transactional and query capabilities are limited to data residing inside a relational database whereas text and multimedia data residing at a web site are only viewed by the use of Hyper Text Markup Language (HTML). The World Wide Web was originally built for human consumption, and although everything on it is machine-readable, everything is not machine-understandable. It is hard to automate anything on the web, and because of the volume of information the web contains, it is not possible to manage it manually.

W3C (web address http://www.w3c.org) is an international industry consortium to lead the World Wide Web to its full potential by developing common protocols that promote its evolution and ensure its interoperability. The solutions so far proposed by W3C in Extensible Markup Language (XML) and Resource Description Framework (RDF) incorporate metadata to describe the data contained on the web. Metadata is "data about data" or specifically "data describing Web resources" in the context of the World Wide Web. The distinction between "data" and "metadata" is not an absolute one and it is a distinction created primarily by an application. Programs and autonomous agents can gain knowledge about data from metadata specifications.

The RDF model draws well-established principles from various data representation communities. RDF properties may be thought of as attributes of resources and in this sense correspond to traditional attribute-value pairs. The basic model consists of three object types,:

(1) Resources: All things being described by RDF expressions are called resources. A resource may be an entire Web page; for example the HTML document http://www.w3.org/Overview.html. A resource may be a part of a Web page; e.g. a specific HTML or XML element within a document source. A resource may also be a whole collection of pages; e.g. an entire Web site. Resources are identified by universal resource identifiers or URIs. Anything can have URI; the extensibility of URIs allows the introduction of identifiers for any imaginable entity.

(2) Properties: A property is a specific aspect, characteristic, attribute or relation used to describe a resource. Each property has a specific meaning, defines its permitted values, the types of resources it can describe, and its relationship with other properties.

(3) Statements: A specific resource together with a named property plus the value of that property for that resource is a RDF statement. These three individual parts of a statement are called, respectively, the subject, the predicate and the object. The object of a statement (i.e. the property value) can be another resource or it can be a literal, i.e. a resource (specified by a URI) or a simple string or other primitive data type defined by XML.

A simple example statement "John Doe is the creator of the resource http://www.w3.org/home/John" has the Subject (resource) http://www.w3.org/home/John, Predicate (property) as "Creator" and Object (literal) as "John Doe". Meaning in RDF is expressed through reference to a schema. A schema is a place where definitions and restrictions of usage for properties are documented. In order to avoid conflicts in definitions of the same term, RDF uses the XML namespace facility where a specific use of a word is tied to the dictionary (schema) where the definition exists. Each predicate used in a statement must be identified with exactly one namespace, or schema. RDF model also allows qualified property value where the object of the original statement is the structured value and the qualifiers are further properties of a common resource. To represent a collection of resources, RDF uses an additional resource that identifies the specific collection. This resource should be declared to be an instance of one of the container object types, namely,

(1) Bag (an unordered list of resources or literals),

(2) Sequence (an ordered list of resources or literals) and

(3) Alternative (a list of resources or literals that represent alternatives for the single value of a property).

A common use of containers is the value of a property. When used in this way, the statement still has a single statement object regardless of the number of members in the container; the container resource itself is the object of the statement.

Use of metadata was so far popular in relational databases to describe attributes, number and types of columns in tables, foreign-key/primary-key relationships, views etc. in a relational schema. SQL (Structured Query Language) queries made against a relational schema are resolved by fetching metadata from data dictionaries (or repository for metadata definitions) to interpret data fetched from data files during execution of a relational operation. Query executions are independent of any application domain specific features. In a similar manner, Resource Description Framework (RDF) is a foundation for representing and processing metadata and data for the World Wide Web; it provides interoperability between applications that exchange machine-understandable information on the web. The broad goal of RDF is to define a mechanism for describing resources that makes no assumptions about a particular application domain, nor defines the semantics of any application domain. RDF relies on the support of XML (extensible markup language) and its model resembles an entity-relationship diagram. In object-oriented design terminology, resources correspond to objects and properties correspond to instance variables. To facilitate the definition of metadata, RDF represents a class system much like object-oriented programming and modeling systems. A collection of classes is called a schema. Schemas may themselves be written in RDF.

Representation of "data about data" (metadata) to achieve application independent interoperable solutions carries the basic similarity between relational databases and RDF. However, RDF does not carry facilities for specifying queries making use of metadata, so far possible in a relational database. Query capabilities enable users to construct arbitrary types of data on the fly for application processing logic to apply. Additionally, relational databases are having advanced capabilities in universal servers to specify application interfaces embedded inside SQL query expressions to represent operations or methods to apply over constructed data. Such important possibilities are also missing from RDF. Lack of such facilities is the limitation of RDF to address evolving electronic business needs in its completeness.

Relational algebra incorporates algebraic operations like join, select, project, union, intersection etc. Such operations are expressed in queries against a relational schema. As opposed to this scenario, web entities are accessed by navigation through Uniform Resource Identifiers (URIs). An amalgamation of these two paradigms is the desired goal to achieve in electronic business. Relational operations over RDF definitions for resources and their attributes are possible exploiting relationships over resources and structured property values and normalizing them in a back end relational database. Queries involving join, select, project and other relational operations can be effectively used to extract desired values and properties of resources. Without such a mechanism, web surfing in conjunction with complex automated business to business services and transactions are not possible.

Electronic commerce and services have introduced many new ways of trading allowing interaction between groups that previously could not economically afford to trade among one another. Whereas previously commercial data interchange involved mainly the movement of data fields from one computer to another, the new model for web-based commerce and services is typically dependent on intelligent processing and interactions for the transactions to take place. This involves understanding and specifying business concepts represented in the interchanged data and subsequent application of business-specific rules or methods to the interchanged data. Transactional and query facilities with embedded method interfaces can lead to such a powerful scheme. Query specifications with embedded interfaces are currently present in object relational databases or universal servers. Object relational databases with business logic bound inside the server offer distinct directions for resolving similar complex issues over XML/RDF definitions and Java classes. Transactional and query facilities to an object relational database are possible through thin client windows incorporating a persistent connectivity with the database. Persistent connectivity to a database system is not possible in a simple browser for stateless web navigation.

XML/RDF documents are interchanged based upon HTTP (Hypertext transfer protocol) which is different from IIOP (Internet inter ORB protocol). HTTP is the main communication mechanism among web browsers and servers. It is a stateless protocol implying that there is no way for the client and the server to know each others state. Since web is stateless, each transaction consumes time and resources in the setup and closing of network and database connections. For large transaction processing applications, this overhead will be significant. Internet inter ORB protocol (IIOP) is a dynamic invocation interface for the web. This protocol maintains the session between the client and the server objects until either side disconnects. It provides persistent connectivity over the web for distributed objects. The OMG (Object Management Group with web address http://www.omg.org) is an industry consortium to create a component based software marketplace by establishing industry guidelines and detailed object management specifications to provide a common framework for application development. Common Object Request Broker Architecture (CORBA) from OMG specifies the Object Request Broker (ORB) that allows applications and programs to communicate with one another no matter where they reside on the web. The IIOP specification defines a set of data formatting rules, called CDR (Common Data Representation) which is tailored to the data types supported in the CORBA interface definition language (IDL). Electronic business transactions and query servers implementing structured query language (SQL) processing engine require internet protocols for document transfers as well as object executions with persistent connectivity over the web. As a result, such an engine must build on top of both HTTP and IIOP. Traditional browsers for navigation need to be augmented with additional capabilities for occasional creation, maintenance and destruction of one or more client windows interfacing databases over the web for transactions and collaborations. These windows require IIOP for persistent connectivity.

A database schema can be partitioned over the web in such a way that disparate business logic and business objects can exist with relevant data and views over the web. Unifying the object paradigm and relational model paradigm is the mainstream effort across the industry. Unified model for distributed relational databases integrated with object model is the key to many storage and manipulation issues for the electronic business. Universal relational database servers are available from different database vendors to offer general extensibility and features for electronic business. One can extend types of attributes in tables and integrate routines defined by users written in high level programming languages. Such products offer the facilities of user-defined routines and packages. A user-defined routine (UDR) is a routine that a user creates and registers in the system catalog tables and that is invoked within a SQL statement or another routine. A function is a routine that optionally accepts a set of arguments and returns a set of values. A function can be used in SQL expressions. A procedure is a routine that optionally accepts a set of arguments and does not return any values. A procedure cannot be used in SQL expressions because it does not return a value. An UDR can be either a function or a procedure. The ability to integrate userdefined routines, packages and functions within SQL is the extensibility feature offered by universal servers and such features are useful for electronic business.

Uniform Resource Identifiers are frequently embedded in XML and HTML pages where a browser can navigate through a resource identifier to find and manipulate web objects. A resource can also identify an object relational schema component over the web. RDF documents represent metadata that could be directly derived from one or more object relational database(s). Information existing inside object relational databases presented in XML/RDF definitions makes an information hierarchy over the web that should be seamlessly navigated and queried. This kind of seamless interoperability can prove to be very valuable in electronic business and commerce. However, these possibilities are not present in current state of the art.

As described above, there is a clear need in the art for automated web functionality in electronic business over the information stored and exchanged across the internet, requiring (a) to support a generic execution model for arbitrary type construction from XML/RDF definitions followed by business logic execution anywhere on the web, (b) to support uniform object SQL query facilities over a single virtual database unifying multiple object relational databases viewed by related XML/RDF documents and (c) to support uniform navigational as well as transactional facilities over data and metadata definitions extending capabilities in traditional browsers for occasional creation and maintenance of thin client windows for persistent connectivity with remote databases. There are further needs in XML/RDF processing framework to unify and support various techniques for buyer/seller relationships in electronic commerce, (a) by allowing navigation through web pages (XML/RDF definitions) for inspecting and implicitly generating object SQL queries, (b) by executing such queries in one or more object relational engines over the internet to generate further web pages (XML/RDF definitions) with more accurate or elaborate information, (c) by seamlessly repeating such processes described in (a) and (b) if necessary and (d) by performing transactions on focused data items anywhere on the web.

SUMMARY OF THE INVENTION

The present invention solves the aforementioned deficiencies of the prior art and solves other problems that will be understood and appreciated by those skilled in the art upon reading and understanding the present specification. It is a primary objective of the present invention to provide a mechanism in Extensible Markup Language (XML) and Resource Description Framework (RDF) for representing and navigating higher level specifications for data/metadata and for constructing arbitrary types by SQL queries with embedded method interfaces. Specifications for SQL queries with embedded business application logic could be either represented in XML/RDF documents or such queries for transactions could be triggered through thin client windows communicating persistently with remote databases. These object SQL queries apply uniformly to one or more object relational databases over the web to manipulate data and to construct further XML/RDF documents for navigation and inspection. This way, a uniform paradigm for navigation through Uniform Resource Identifiers within XML/RDF documents and querying one or more object relational schema components identified seamlessly by Uniform Resource Identifiers over the web will evolve, thereby addressing various needs in electronic business and commerce.

In one embodiment of the invention, a virtual unified Database over multiple object relational databases over the web is described. Application business logic, messaging services and object request brokers reside inside an object relational database server eliminating the need for a middle tier. Uniform Resource Identifiers are stored in table columns of relational databases as locators of elements inside component relational schema with Java classes distributed over the net. Java classes encapsulate or package business logic to be applied on relational data or other multimedia data. Types defined by users to represent attributes in tables involve packaged object definitions to encapsulate methods or operations. These operations are applied through Object Request Brokers. In this invention, user-defined package contains interface definitions (call specifications) where an explicit mapping of data item types to tables and attributes are made to ensure safe environment of method application. Such definitions of interfaces inside packages are used to apply methods over records constructed by relational operations and these interfaces are embedded within object SQL queries. These interfaces are implemented as methods in Java classes with appropriate mappings to Java argument types. Therefor a uniform paradigm for multi-tier client/server without a middle tier application server is presented.

In another embodiment, XML/RDF definitions to represent data and metadata are mapped to relational descriptions for records and their relationships through foreign keys and primary keys. Uniform Resource Identifiers are unique identifiers for entities over the web and in this invention these identifiers are used as primary key and foreign key values in a relational model of description. A relationship expressed in RDF description of resources can be directly mapped to the primary key/foreign key relationship of records in a relational database. A collection of resources is defined by RDF container types; namely Bag, Sequence and Alternative. These container types can be normalized into proper foreign key and primary key relationships so that SQL queries involving join and other operators may be able to construct records with defined container relationships. Multi-part primary keys for sequences and different normal forms in relational model capture various semantics possible in RDF data/metadata definitions. Thus multiple related XML/RDF documents distributed over the web are mapped to data and metadata in multiple related object relational databases for object SQL queries to fetch and manipulate information.

In another embodiment, method interfaces are represented as virtual attributes with specifications for parameters as resources and collection of resources. Definitions for methods are given in XML namespace schema definition. This invention further extends the scope of representing SQL queries in XML/RDF specifications by defining XML namespace for relational operator definitions. All the key words in SQL are defined in namespace with respective meanings. SQL queries are implicitly constructed from sets of XML/RDF definitions in semantically related documents. This invention further represents object SQL queries by having method interfaces embedded within queries in XML/RDF format. This way a specific XML/RDF document can be carrying an object SQL query for execution within one or more object relational schema components over the web.

In yet another embodiment, XML/RDF documents with data/metadata definitions constitute views over databases. A part of a database or union over several databases can be viewed by a document. All database updates are viewed by generating new documents replacing old ones. A view or a union over views or a full database can be updated by creating client windows maintaining persistent links through object brokers. Such windows are maintained as long as transactions persist. Transactions over views or a database cannot be done without such a client window. Only read-only implicit queries are allowed without such client windows. Object SQL queries uniformly manipulate local and/or remote database table, attributes and objects. Querying and viewing multiple object relational schema components over the web can also include legacy and other large existing database systems.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a diagram illustrating various information sources over the internet relevant to the present invention.

FIG. 2 is a diagram illustrating buyers/suppliers in an extende