Apparatus for automatically generating source code

• US 7,197,739 B2
• Filed: 01/30/2001
• Issued: 03/27/2007
• Est. Priority Date: 01/31/2000
• Status: Active Grant

• Current Assignee: British Telecommunications Public Limited Company
• Original Assignee: British Telecommunications PLC (BT Group PLC)
• Inventors: Preston, Keith R, Leathem, Caroline A
• Primary Examiner(s): Zhen; Wei
• Assistant Examiner(s): DENG, ANNA CHEN
• Application Number:
  US10/169,352
• Publication Number:
  US 20030046061A1
• Time in Patent Office: 2,247 Days
• Field of Search: 717/104, 717/109, 717/113, 717/136, 717/139, 717/143, 717/144, 706/23, 706/47
• US Class Current: 717/106
• CPC Class Codes: G06F 8/30 Creation or generation of source code
• CPC Class Codes: G06F 8/436 Semantic checking

Abstract

Software is automatically generated from one or more predefined functions in accordance with an input statement entered in natural language. Semantically meaningful elements are extracted from the input statement and one or more sets of second semantically meaningful elements are extracted from the predefined functions. At least one of a condition, an action and/or a statement is identified in the input statement. The first semantically meaningful elements are compared with the second semantically meaningful elements to identify predefined functions that correspond to action(s) and/or statement(s) of the input statement. At least some of the first semantic elements are combined in accordance with identified conditions to generate corresponding condition variables. Identified functions and condition variables are combined according to a set of predetermined rules to generate software.

Claims

1. A method of automatically generating software from one or more predefined functions in accordance with an input statement entered in natural language, the method comprising:

(i) analyzing an input statement for semantic content to extract first semantically meaningful elements using an analysis algorithm;

(ii) analyzing one or more predefined functions for semantic content to extract one or more sets of second semantically meaningful elements using the same analysis algorithm as in step (i);

(iii) identifying at least one of a condition, an action and/or a statement in the input statement;

(iv) comparing the first semantically meaningful elements with the second semantically meaningful elements to identify one or more predefined functions that correspond to one or more action and/or statement of the input statement;

(v) combining at least some of the first semantic elements in accordance with any conditions identified at step (iii) to generate corresponding condition variables;

(vi) combining functions and condition variables identified at steps (iv) and (v) according to a set of predetermined rules to generate software.

2. A method as in claim 1, in which comparing step (iv) includes:

a) inputting a set of second semantically meaningful elements into a predetermined rule;

b) inputting the first semantically meaningful elements into the rule;

c) processing the rule;

d) evaluating the result of step (c); and

e) repeating steps (a) to (c) for different sets of second semantically meaningful elements until as solution is evaluated at step (d).

3. A method as in claim 1, in which step (v) includes for each condition identified at step (iii):

a) identifying an event governing the condition;

b) identifying an entity associated with the event;

c) identifying a relation between the event and the entity; and

d) concatenating the event, the entity and the relation therebetween, thereby forming a condition variable.

4. A method as in claim 1, in which the functions and condition variables identified at steps (iv) and (v) are inserted respectively into a predetermined code template.

14. A method as in claim 1 wherein said predefined functions are also defined using natural language.

5. Computer apparatus for automatically generating software from one or more predefined functions in accordance with an input statement entered in natural language, the apparatus comprising:

(i) extracting means for extracting first semantically meaningful elements from the input statement and for extracting one or more sets of second semantically meaningful elements from the one or more predefined functions using a common extraction algorithm for both said extraction processes;

(ii) identifying means for identifying conditions, actions and/or statements in the input statement;

(iii) comparing means for comparing said first semantically meaningful elements with said second semantically meaningful elements to identify one or more predefined functions that correspond to an action and/or statement in the input statement;

(iv) first combining means for combining at least some of the first semantic elements in accordance with conditions identified by the identifying means to generate corresponding condition variables;

(v) second comparing means for combining the condition variables and predefined functions according to a set of predetermined rules to generate software.

6. Computer apparatus as in claim 5, wherein the extracting means is a natural language parser.

7. Computer apparatus as in claim 5, including a predetermined code template.

8. Computer apparatus as in claim 5, wherein the identifying means identifies conditions, actions and statements in accordance with the occurrence of one or more predetermined semantic patterns.

9. Computer apparatus as in claim 5, wherein the software is used in the control of terminal devices in a communications system.

15. Computer apparatus as in claim 5 wherein said predefined functions are also defined using natural language.

10. Computer apparatus for generating code instructions for controlling equipment to carry out one or more activities, said apparatus comprising:

a) an input for receiving instructions for use in controlling the equipment;

b) received instruction processing means for extracting one or more semantically meaningful elements describing operations and extracting one or more semantically meaningful elements describing logical operators from received instructions using a common extraction algorithm for both said extraction processes;

c) means for storing a plurality of code instructions for controlling the equipment to carry out activities;

d) means for processing stored code instructions to identify a code instruction relevant to an operation extracted from a received instruction by identifying extracted elements from the input instructions and comparing the extracted elements with each other;

e) means for translating extracted logical operators into code by identifying extracted elements from the input instructions and comparing the extracted elements with each other; and

f) code generating means for selecting at least one identified code instruction and combining it with at least one translated logical operator to generate said code instructions for controlling the equipment to carry out the one or more activities.

11. Computer apparatus as in claim 10 wherein said received instructions comprise natural language.

12. Computer apparatus as in claim 11 wherein the received instruction processing means comprises a parser.

13. Computer apparatus as in claim 10 wherein the received instruction processing means also provides the means for processing stored code instructions.