API mashup exploration and recommendation

1. A method, comprising:

collecting a plurality of computer-readable source code from an application programming interface (API) repository;

performing API call code slicing on the computer-readable source code to identify API call codes from the computer-readable source code, the API call code slicing comprising, for an API call code comprising multiple code files with multiple API calls within a single project:

collecting a plurality of Uniform Resource Locator (URL)s in the multiple code files;

filtering the plurality of URLs to identify a plurality of host URLs; and

filtering the multiple code files to identify code files which include the plurality of host URLs and which include two or more URLs;

extracting an API host and an endpoint for each of the API call codes identified in the API call code slicing;

compiling an API mashup database based on the extracted API host and endpoints, the API mashup database including a plurality of API mashups wherein each respective API mashup:

relates to a corresponding software program;

includes a plurality of respective API's used by the corresponding software program;

indicates a corresponding host and endpoint for each respective API; and

indicates a usage relationship between the respective API's of the respective API mashup; and

providing a recommended API mashup of the API mashup database in a user interface.

2. The method of claim 1 , wherein the API repository comprises GITHUB®.

3. The method of claim 1 , wherein collecting the plurality of computer-readable source code comprises:

identifying a plurality of API call methods for a programming language;

performing a code search operation on the API repository using the identified API call methods of the programming language; and

filtering the results of the code search operation to identify code files including two or more API calls.

4. The method of claim 3 , wherein collecting the plurality of computer-readable source code further comprises:

identifying single code files in the results of the code search operation including two or more API calls; and

identifying multiple code files for a single project in the results of the code search operation including multiple API calls.

5. The method of claim 1 , wherein the API call code slicing comprises:

for an API call code comprising a single code file identified as including two or more API calls:

traversing an abstract syntax tree of the single code file to locate the API call code in the single code file; and

removing code statements in the single code file which are irrelevant to the API call code.

6. The method of claim 1 , further comprising:

receiving user-defined search parameters for a recommended API mashup from a user,

wherein the recommended API mashup is further recommended based on the user-defined search parameters.

7. The method of claim 6 , wherein the user-defined search parameters are selected from a group including host URL, endpoint, API keyword, and API repository.

8. One or more non-transitory computer-readable media containing instructions, which, when executed by one or more processors, cause a system to perform operations, the operations comprising:

collecting a plurality of computer-readable source code from an application programming interface (API) repository;

performing API call code slicing on the computer-readable source code to identify API call, the API call code slicing comprising, for an API call code comprising multiple code files with multiple API calls within a single project:

collecting a plurality of Uniform Resource Locator (URL)s in the multiple code files;

filtering the plurality of URLs to identify a plurality of host URLs; and

filtering the multiple code files to identify code files which include the plurality of host URLs and which include two or more URLs;

extracting an API host and an endpoint for each of the API call codes identified in the API call code slicing;

compiling an API mashup database based on the extracted API host and endpoints, the API mashup database including a plurality of API mashups wherein each respective API mashup:

relates to a corresponding software program;

includes a plurality of respective API's used by the corresponding software program;

indicates a corresponding host and endpoint for each respective API; and

indicates a usage relationship between the respective API's of the respective API mashup; and

providing a recommended API mashup of the API mashup database in a user interface.

9. The computer-readable media of claim 8 , wherein the API repository comprises GITHUB®.

10. The computer-readable media of claim 8 , wherein collecting the plurality of computer-readable source code comprises:

identifying a plurality of API call methods for a programming language;

performing a code search operation on the API repository using the identified API call methods of the programming language; and

filtering the results of the code search operation to identify code files search including two or more API calls.

11. The computer-readable media of claim 10 , wherein collecting the plurality of computer-readable source code further comprises:

identifying single code files in the results of the code search operation including two or more API calls; and

identifying multiple code files for a single project in the results of the code search operation including multiple API calls.

12. The computer-readable media of claim 8 , wherein API call code slicing comprises:

for an API call code comprising a single code file identified as including two or more API calls:

traversing an abstract syntax tree of the single code file to locate the API call code in the single code file; and

removing code statements in the single code file which are irrelevant to the API call code.

13. The computer-readable media of claim 8 , the operations further comprising:

receiving user-defined search parameters for a recommended API mashup from a user,

wherein the recommended API mashup is further recommended based on the user-defined search parameters.

14. The computer-readable media of claim 13 , wherein the user-defined search parameters are selected from a group including host URL, endpoint, API keyword, and API repository.

15. A system comprising:

one or more processors; and

one or more non-transitory computer-readable media containing instructions, which, when executed by one or more processors, cause a system to perform operations, the operations comprising:

collecting a plurality of computer-readable source code from an application programming interface (API) repository;

performing API call code slicing on the computer-readable source code to identify API call codes from the computer-readable source code, the API call code slicing comprising, for an API call code comprising multiple code files with multiple API calls within a single project:

collecting a plurality of Uniform Resource Locator (URL)s in the multiple code files;

filtering the plurality of URLs to identify a plurality of host URLs; and

filtering the multiple code files to identify code files which include the plurality of host URLs and which include two or more URLs;

extracting an API host and an endpoint for each of the API call codes identified in the API call code slicing;

compiling an API mashup database based on the extracted API host and endpoints, the API mashup database including a plurality of API mashups wherein each respective API mashup:

relates to a corresponding software program;

includes a plurality of respective API's used by the corresponding software program;

indicates a corresponding host and endpoint for each respective API; and

indicates a usage relationship between the respective API's of the respective API mashup; and

providing a recommended API mashup of the API mashup database in a user interface.

16. The system of claim 15 , wherein collecting the plurality of computer-readable source code comprises:

identifying a plurality of API call methods for a programming language;

performing a code search operation on the API repository using the identified API call methods of the programming language; and

filtering the results of the code search operation to identify code files search including two or more API calls.

17. The system of claim 16 , wherein collecting the plurality of computer-readable source code further comprises:

identifying single code files in the results of the code search operation including two or more API calls; and

identifying multiple code files for a single project in the results of the code search operation including multiple API calls.

18. The system of claim 15 , wherein the operations further comprise:

receiving user-defined search parameters for a recommended API mashup from a user,

wherein the recommended API mashup is further recommended based on the user-defined search parameters.