Method for automatic extraction of design from standard source code
A system and method for automatic code-design and file/database-design association. Existing source code is analyzed for process and control elements. The control elements are encapsulated as augmented state machines and the process elements are encapsulated as kernels. The new elements can then have meta-data attached (including, a name, I/O method, and test procedures), allowing software code sharing and automatic code/file/database upgrading, as well as allowing sub-subroutine level code blocks to be accessed directly.
1. A computer-implemented method for automatic association and testing of candidate code within a candidate code database to software design code comprising:
decomposing the software design code into McCabe code blocks each consisting of a transformation process;
generating a list of keywords and test procedures for each of the McCabe code blocks;
matching the keywords of a selected one of the McCabe code blocks to keywords of the candidate code database to determine the candidate code blocks matching said selected McCabe code blocks;
executing the test procedures on the matching candidate code blocks to determine at least one candidate code block that can be used with the selected one of the McCabe code blocks.
2. The computer-implemented method of claim 1 , wherein the list of keywords comprises at least one of database type and database schema; and wherein the step of matching includes identifying potential candidate code databases based upon at least one of database type and database schema.
3. The computer-implemented method of claim 1 , the test procedures comprising input and output variables and expected results from executing the selected McCabe code block with the input and output variables.
4. The computer-implemented method of claim 1 , further comprising eliminating candidate code blocks that cannot be used with the selected one of the McCabe code blocks.
5. The computer-implemented method of claim 1 , further comprising associating one of the candidate code blocks with the selected one of the McCabe code blocks.
6. The computer-implemented method of claim 5 , further comprising identifying the associated candidate code block based upon goals to be achieved by the software design code.
7. The computer-implemented method of claim 6 , wherein the goals include one or more of maximum input dataset size, maximum system resource utilization, maximum performance, maximum Amdahl scaling, maximum accuracy, and minimum price.