GB2445794A

GB2445794A - Generating program code from natural language descriptions

Info

Publication number: GB2445794A
Application number: GB0700950A
Authority: GB
Inventors: Ian Keith Hamilton
Original assignee: Individual
Current assignee: Individual
Priority date: 2007-01-18
Filing date: 2007-01-18
Publication date: 2008-07-23
Also published as: GB0700950D0

Abstract

Computer cod is generated from a set of pre-defined system tokens each of which has a natural language label. A process builder can be used to generate program code by sequencing tokens and component references, previously created to describe aspects of the system requirements. The sequence data may also be stored so that it can be reloaded to generate the code later.

Description

Field of the invention.

This invention relates to process of generating program source code, within the field of computer software development.

Background.

Traditionally all computer programs were developed by programmers writing code by hand. There have been a number of attempts to automate parts of the process, in order to speed it up. Most relate to object oriented design methods, generating some parts of the code required, based on object or class descriptions. These processes, 00 and other, all share one common aspect: that they are designed as an aid to the programmer and intended to generate only part of the source code required. The programmer still has to complete the code by hand, in particular the processes that represent the business logic, which is frequently complex and time consuming. It is also necessary for the business representative to have explained it to the programmer beforehand. The body of this invention covers the task of automatically generating the individual application processes as code from such user defined representations as might be presented to a programmer, in order that (s) he might be able to complete the task.

Description of the invention.

The system contains a set of pre-defined system tokens. Each token represents a specific action, or part of an action, which the process builder can recreate as units of program source code in any supported programming or scripting language, depending on the inclusion context. Each token also has a natural language label, or representation, so that when sequenced correctly, the token labels and other inclusions, taken in order, will meaningfully describe, or represent, the details of the process. Natural language, in this context, refers to the user's normal, spoken language, such as English or French.

A person with an understanding of the desired outcome, is be able to define their requirements within the system by specifying, in the correct sequence, a number of tokens and links to pre-existing components, with values where appropriate, such that each required process is unambiguously described by said sequence. These sequences hold the information necessary in order to generate the computer software processes and are stored as a collection of these tokens.

The sequence data created, may be stored in a supported electronic format, designed for the purpose and that data storage may be in the form of binary files, XML files, other file types or a database; it is not important. Where the specification data is stored, it must first be reloaded into memory, prior to code generation. The system may alternatively generate computer source code directly, without ever saving the sequence data to file.

The process builder method reads the sequences representing the process detail and generates computer program source code in a supported programming language, based on the selected components, tokens and values defining the process. Figure 1 shows the logical steps of the process code generator. The source code generated by this process can then integrated with any other code, components or processes required to form the whole application. * 2

Claims

Claims What is claimed is: 1. A method for generating computer source

code ready for compilation it into applications, application fragments or scripting pages, comprising: a collection of tokens representing stages in process flow; a method of collecting user requirements as a sequence of said tokens, links to system components and fixed values; a process of generating source code from the user defmed collections of tokens and links.
2. The method of claim 1, wherein said step of collecting user requirements requires the existence of a set of tokens representing stages, actions or partsof a process.
3. The method of claim 1, wherein said step of collecting user requirements further comprises presenting one or more appropriate data capture screens.
4. The method of claim 1, wherein said step of generating source code from the user defined collections of tokens, links and values will produce computer source code by iterating through the sequence, adding the corresponding code and values associated with each item in the sequence.