Automated code-generation for cross-language development, testing, and integration

• US 9,778,922 B2
• Filed: 10/28/2015
• Issued: 10/03/2017
• Est. Priority Date: 10/31/2014
• Status: Active Grant

• Current Assignee: BAE Systems Information and Electronic Systems Integration Incorporated (BAE Systems Plc)
• Original Assignee: BAE Systems Information and Electronic Systems Integration Incorporated (BAE Systems Plc)
• Inventors: Hill, III, Edward H.
• Primary Examiner(s): Vo, Ted T
• Application Number:
  US14/925,119
• Publication Number:
  US 20160124731A1
• Time in Patent Office: 706 Days
• Field of Search: 717106-109, 717135, 717137
• CPC Class Codes: G06F 8/31 Programming languages or programming paradigms
• CPC Class Codes: G06F 8/47 Retargetable compilers
• CPC Class Codes: G06F 8/51 Source to source

Abstract

A system and method provide for easy sharing of data between different software languages. A method begins by creating a definition defining a data structure with a domain specific language. The definition is then input to a code generator which generates data structures and algorithms in a first software language. The same generator software also creates equivalent data structures and algorithms in a second software language that is different than the first software language. The two output implementations provide compatible utilities for marshalling and de-marshalling data back and forth between the first software language and the second software language without requiring further manipulation of the two implementations.

Claims

1. A method comprising:

defining, by a processor, a set of user-defined data types with an encoding;

using the encoding and a code-generator utility to generate, by the processor, a first at least one class to implement each of the user-defined data types within the set of user-defined data types in a first software general-purpose language (GPL) as a first container;

using the encoding and the code-generator utility to generate, by the processor, a second at least one class to implement each of the set of user-defined data types within the set of user-defined data types in a second software GPL as a second container;

running a code-generator to generate, by the processor, software configured to marshal each of the set of user-defined data types within the set of user-defined data types in the first and second at least one classes;

marshalling, by the processor, data directly between the first software language and the second software language utilizing a Matlab executable (MEX) interface; and

de-marshalling, by the processor, the marshalled data directly into the second software language utilizing the MEX interface.

2. The method of claim 1 further comprising:

using the encoding to generate utilities in the first software language and to generate utilities in the second software language.

3. The method of claim 2 wherein the utilities further comprise:

a marshalling utility in the first software language configured to marshal data between the first container of the first software language and the second container of the second software language; and

a de-marshalling utility in the second software language configured to de-marshal data into the second software language.

4. The method of claim 3 further comprising:

using the marshalling utility and de-marshalling utility to marshal and de-marshal data from the first software language directly to the second software language.

5. The method of claim 1 further comprising:

capturing simulation data representing the state of a simulation test in the first software language at a point in time;

marshalling the simulation data directly to the second software language to create marshalled simulation data;

de-marshalling the marshalled simulation data directly into the second software language; and

viewing a state of the second software language at the point in time.

6. The method of claim 5 further comprising:

running the simulation test from the point in time in the second software language to a second different point in time.

7. The method of claim 5 wherein the marshalling further comprises:

marshalling the simulation data into a compacted binary format.

8. The method of claim 1 wherein the encoding utilizes, at least in part, JavaScript Object Notation (JSON).

9. The method of claim 1 wherein the first at least one class comprises one or more of the group of:

signed integers, unsigned integers, float values, double float values, complex float values, complex double float values, Booleans, strings, dynamic-sized ordered arrays of primitive types, code-generated types, dynamically-sized ordered arrays, associative arrays, fixed-length arrays, and graphs.

10. The method of claim 1 wherein the first software GPL is one of the group of:

Matlab and Octave.

11. The method of claim 1 wherein the defining a set of user-defined data types further comprises:

defining a set of user-defined data types that has multiple levels of hierarchy.

12. The method of claim 1 wherein the second software GPL is one of the group of:

C, C++ and CUDA.

13. The method of claim 1, wherein a correspondence between each of the set of user-defined data types within the set of user-defined data types and the first and second at least one classes in the first software language and the second software language is 1:

1.

14. A system comprising:

a processor;

a data type generation logic configured to cause the processor to allow a user to define a set of user-defined data types;

a code generation logic configured to cause the processor to generate a first at least one class to implement each of the set of user-defined data types within the set of user-defined data types in a first container of a first software general-purpose language (GPL), and wherein the code generation logic is configured to cause the processor to generate a second at least one class to implement each of the set of data types within the set of user-defined data types in a second container of a second software GPL;

wherein the code generation logic is further configured to cause the processor to generate software configured to marshal each of the set of user-defined data types within the set of user-defined data types in the first and second at least one classes;

a data marshalling logic configured to cause the processor to marshal data directly between the first software language and the second software language utilizing a Matlab executable (MEX); and

a data de-marshalling logic configured to cause the processor to de-marshal the marshalled data directly into the second software language utilizing the MEX interface.

15. The system of claim 14 wherein the processor further comprises:

an encoding configured to allow the user to define at least a portion of the set of user-defined data types.

16. The system of claim 14 wherein the marshalling logic is configured to cause the processor to generate marshalling utilities in the first software language configured to be used to capture state and data values of the first software language at a point in time;

and

wherein the de-marshalling logic is configured to cause the processor to generate de-marshalling utilities in the second software language configured to be used to directly marshal and de-marshal the state and data values of the first software language into the second software language allowing the second software language to be run from the point in time.

17. The system of claim 14 further wherein the set of user-defined data types has a plurality of levels of hierarchy.

18. The system of claim 14 wherein the first software GPL is one of the group of:

Matlab and Octave.

19. The system of claim 14 wherein the second software GPL is one of the group of:

C C++ and CUDA.

20. The system of claim 14, wherein a correspondence between each of the set of user-defined data types within the set of user-defined data types and the first and second at least one classes in the first software language and the second software language is 1:

1.