Method and apparatus for converting legacy programming language data structures to schema definitions

• US 20040044678A1
• Filed: 12/18/2002
• Published: 03/04/2004
• Est. Priority Date: 08/29/2002
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Fang, Sheng, Kalia, Suman K., Green, John H.
• Granted Patent: US 7,533,102 B2
• Time in Patent Office: Days
• US Class Current: 707/102
• CPC Class Codes: G06F 16/284 Relational databases
• CPC Class Codes: G06F 8/31 Programming languages or programming paradigms

Abstract

A method and apparatus for creating a schema definition from a language-specific metamodel of a data structure written in a legacy computer language involves walking through the nodes of the metamodel, and creating corresponding schema element objects in a schema tree that is serialized to create the schema definition. The creation of the schema element objects follows rules, and permits the definition of nested data structures having conjunctive-type and disjunctive-type membership.

Claims

1. A method for generating a schema definition, comprising:

receiving a language-specific metamodel generated from a nested data structure definition written in a legacy computer language;

naming and defining a complex type schema element for the nested data structure; and

creating the schema definition by walking through the metamodel to examine each node in the metamodel, and for each node, creating a respective schema element according to a predefined mapped relationship.

2. A method as claimed in claim 1 wherein receiving comprises receiving a parse tree generated from the nested data structure definition written in one of C, PL/I, high level assembler and Cobol.

3. A method as claimed in claim 1 wherein naming and defining comprises setting a content model of the complex type schema element to sequence.

4. A method as claimed in claim 3 wherein creating the schema definition further comprises:

populating a schema tree; and

serializing the schema tree after it is populated.

5. A method as claimed in claim 4 wherein populating a schema tree comprises:

selecting a first/next node in the walk through the language-specific metamodel, in accordance with a breadth-first selection algorithm;

examining the elements of the node to determine a type of the elements; and

applying the predefined mapped relationship to the node in accordance with the node'"'"'s type to generate a schema element.

6. A method as claimed in claim 5 wherein applying the predefined mapped relationship comprises:

generating the schema element by creating an object of a type determined by the mapped relationship;

specifying a name for the object; and

setting all required attributes of the object of the determined type.

7. A method as claimed in claim 6 wherein specifying a name comprises deriving a name for the object from a list of names of the node, and the ancestors of the node, in accordance with a naming convention.

8. A method as claimed in claim 6 wherein generating the schema element comprises creating one of:

a group schema element;

a simple type schema element; and

a complex type schema element, in dependence upon a type of the node and a type of its elements.

9. A method as claimed in claim 7 wherein setting all required attributes comprises:

setting a content model of sequence or choice for all complex type and group schema elements; and

setting a data type for simple type schema elements.

10. A method as claimed in claim 9 wherein setting a data type comprises applying a mapping from language-specific data types to schema data types.

11. A method as claimed in claim 10 wherein setting a data type further comprises setting optional parameters that apply restrictions on the simple type schema element in accordance with a native interpretation of the node.

12. A method as claimed in claim 10 wherein setting a data type further comprises enclosing an annotation to supply an expected, initial or default value for the node, in accordance with a native interpretation of the node.

13. A method as claimed in claim 6 wherein examining comprises determining if the node comprises at least one union element, and, if so, applying the predefined mapped relationship comprises:

generating a grouping schema element by creating an object that is a child of the node'"'"'s schema element, the child object being a group schema element of choice composition kind; and

for each union element, creating a respective object that is a child of the grouping schema element.

14. A method as claimed in claim 6 wherein examining comprises determining if the node is an array, and if so, the step of applying the predefined mapping comprises:

creating a simple type object;

setting the name of the object;

setting a minimium and a maximum number of fields in the array;

setting a name of the field; and

setting a data type of the field.

17. A method as claimed in claim 9 wherein receiving further comprises receiving a language-specific metamodel containing parts generated from a plurality of nested data structures, and each of these parts, in turn is used to create a respective schema sub-tree having a complex type root that is a child of the root of the schema tree.

18. A method as claimed in claim 17 further comprising receiving user input prior to naming and defining, the user input:

supplying a user-defined name for one of the schema sub-trees that replaces a default name for a complex type schema element that serves as a root of the one of the schema sub-trees;

a list of the parts that are to be used to create respective schema sub-tress; and

a request for a grouping of the parts to form a user-structured schema.

19. A method as claimed in claim 18 wherein receiving user input further comprises receiving a request for a grouping of possible outputs from a system corresponding to nested data structures defined by respective parts identified by the user, the method further comprising:

creating a child object of the root of the schema tree to serve as a root of a schema sub-tree;

setting the name of the child object to a user-supplied name for the grouping;

setting a content model of the child object to choice; and

for each of the schema sub-trees defined for the respective identified parts, inserting an annotation into a first data type element, the annotation including a pattern recognition description that uniquely identifies the possible output.

15. A method as claimed in claim 15 wherein setting a minimum and maximum comprises determining if the array is of fixed-length or variable-length, and setting the number of fields accordingly.

16. A method as claimed in claim 16 wherein setting a data type of the field further comprises determining if the field is non-primitive, and if so, creating a complex type object that is a child of a root of the schema tree, and setting a content model to sequence.

20. Apparatus for generating a schema definition from a definition of a nested data structure written in a legacy computer language comprising:

an interface through which a language-specific metamodel generated from the nested data structure definition is received;

a naming resource that generates names for schema elements derived from names of corresponding nodes of the language-specific metamodel;

a data control system for walking through the nodes of the metamodel; and

a mapping module adapted to receive nodes, inspect their content, and generate a corresponding part of a schema definition to produce a schema definition for the nested data structure.

21. Apparatus as claimed in claim 20 wherein the language-specific metamodel is one of a COBOL metamodel, a C metamodel, a PL/I metamodel and a high level assembler metamodel.

22. Apparatus as claimed in claim 21 wherein the mapping module is adapted to create schema element objects from respective nodes, the schema element objects being arranged to form a schema sub-tree for the nested data structure, and the computer program further comprises a serialization procedure for saving the schema definition to a file.

23. Apparatus as claimed in claim 23 wherein the data control system performs a walk through by selecting successive nodes of the language-specific metamodel according to a breadth-first search algorithm.

24. Apparatus as claimed in claim 24 wherein the mapping module is adapted to inspect the node to determine a type of the node, and then to apply a mapping to the data structure in dependence upon the determined type of the node.

25. Apparatus as claimed in claim 25 wherein the mapping module creates simple type schema elements by assigning a name and data type to the element, and optionally assigning one or more of a data type restriction and an expected default value, in accordance with a native interpretation of the node.

26. Apparatus as claimed in claim 25 wherein the mapping module creates complex type schema elements as child objects of a root node of the schema tree, by assigning a name and a content model of choice or sequence, to the complex type schema elements.

27. Apparatus as claimed in claim 25 wherein the mapping module creates group schema elements by assigning a name, and a content model of choice or sequence, to the group schema elements.

28. Apparatus as claimed in claim 25 wherein the mapping module determines if the node defines an array, in which case a maximum and minimum number of fields in the array, a name of the base field and a type of the base field are included in a simple type schema element definition.

29. Apparatus as claimed in claim 29 wherein the mapping module further determines if the base field is of non-primitive type, in which case it further creates a complex type schema element for the base type.

30. A computer-readable medium containing program instructions for generating a schema definition, comprising:

instructions for receiving a language-specific metamodel generated from a nested data structure definition written in a predetermined legacy computer language;

instructions for naming and defining a complex type schema element for the nested data structure; and

instructions for creating the schema definition by walking through the metamodel to examine each node in the metamodel, and for each node, creating a respective schema element according to a predefined mapped relationship.

31. The computer-readable medium of claim 30 wherein the instructions for receiving comprises instructions for receiving a parse tree generated from the nested data structure definition written in one of C, PL/I, high level assembler and Cobol.

32. The computer-readable medium of claim 30 wherein the instructions for naming and defining comprises instructions for setting a content model of the complex type schema element to sequence.

33. A computer-readable medium of claim 32 wherein the instructions for creating the schema definition further comprises:

instructions for populating a schema tree; and

instructions for serializing the schema tree after it is populated.

34. A computer-readable medium of claim 33 wherein the instructions for populating a schema tree comprises:

instructions for selecting a first/next node in the walk through the language-specific metamodel, in accordance with a breadth-first selection algorithm;

instructions for examining the elements of the node to determine a type of the elements; and

instructions for applying the predefined mapped relationship to the node in accordance with the node'"'"'s type to generate a schema element.

35. The computer-readable medium of claim 34 wherein the instructions for applying the predefined mapped relationship comprises:

instructions for generating the schema element by creating an object of a type determined by the mapped relationship;

instructions for specifying a name for the object; and

instructions for setting all required attributes of the object of the determined type.

36. The computer-readable medium of claim 35 wherein the instructions for specifying a name comprises the instructions for deriving a name for the object from a list of names of the node, and the ancestors of the node, in accordance with a naming convention.

37. The computer-readable medium of claim 35 wherein the instructions for generating the schema element comprises instructions for creating one of:

a group schema element;

a simple type schema element; and

a complex type schema element, in dependence upon a type of the node and a type of its elements.

38. The computer-readable medium of claim 36 wherein the instructions for setting all required attributes comprises:

instructions for setting a content model of sequence or choice for all complex type and group schema elements; and

instructions for setting a data type for simple type schema elements.

39. The computer-readable medium of claim 38 wherein the instructions for setting a data type comprises instructions for applying a mapping from language-specific data types to schema data types.