Intelligent software compiler dependency fulfillment

• US 10,459,701 B2
• Filed: 06/04/2019
• Issued: 10/29/2019
• Est. Priority Date: 11/13/2017
• Status: Active Grant

• Current Assignee: Bank of America Corp.
• Original Assignee: Bank of America Corp.
• Inventors: Singh, Awadhesh Pratap, Jibhe, Dinesh Narendra
• Primary Examiner(s): St Leger, Geoffrey R
• Application Number:
  US16/430,715
• Publication Number:
  US 20190286428A1
• Time in Patent Office: 147 Days
• Field of Search: None
• CPC Class Codes: G06F 11/36 Preventing errors by testing or debugging software
• CPC Class Codes: G06F 16/148 File search processing
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: G06F 8/41 Compilation
• CPC Class Codes: G06F 8/433 Dependency analysis; Data or control flow analysis
• CPC Class Codes: G06N 20/00 Machine learning

Abstract

A machine learning tool for resolving a compiler error in an application is provided. The application and an associated application metadata file may be stored on a server. The machine learning tool may identify one or more referenced external dependencies causing the compiler error. The machine learning tool may comprise a web crawler configured to locate one or more comparable external dependencies. The web crawler may retrieve an external dependent metadata file for each of the located comparable external dependencies and download the comparable external dependent metadata files. The machine learning tool may be configured to compare the metadata of each comparable external dependent metadata file to the metadata of the application metadata file, assign a confidence level relative to a pre-determined confidence level, for each located comparable external dependency, and download the located comparable external dependencies having a confidence level greater than the pre-determined confidence level.

Claims

1. A method for compiling an application, the method comprising:

retrieving from a centralized server;

an application source code associated with the application, the application source code referencing;

a plurality of internal dependencies stored in a cloud, the cloud in synchronization with the centralized server; and

a plurality of external dependencies for compiling the application;

an application metadata file including metadata associated with each of the pluralities of external dependencies; and

performing a plurality of machine learning steps using a machine learning tool, the plurality of steps comprising;

identifying the external dependencies referenced in the application;

instructing a web-crawler, the web-crawler being a component of the machine learning tool, to;

locate a plurality of comparable external dependencies for each of the identified referenced external dependencies, the comparable external dependencies provided by external frameworks on the internet; and

retrieve an external dependent metadata file for each of the plurality of located comparable external dependencies, the external dependent metadata file including metadata associated with the comparable external dependencies;

comparing the metadata of each comparable external dependent metadata file to the metadata of the application metadata file;

assigning, based on the comparison, a confidence level for each comparable external dependency, the confidence level being relative to a pre-determined confidence level, the assigning further comprising, assigning a greater confidence level to one of the located comparable external dependencies having a shorter download time relative to the other located comparable external dependencies;

downloading the located comparable external dependencies for each referenced external dependency having a confidence level greater than the pre-determined confidence level; and

combining and compiling the application source code with the internal dependencies and the downloaded external dependencies.

2. The method of claim 1 wherein, for each referenced external B dependency in the application metadata file and for each of the external dependent metadata files, the metadata includes:

a file name;

a file size; and

a file version.

3. The method of claim 1 wherein the comparing of the B metadata further comprises, comparing a file name, file size and file version of the referenced external dependencies to a file name, file size and file version of the external dependent metadata files for determining the confidence level.

4. The method of claim 1 wherein, when there are a plurality of comparable external dependencies being assigned a confidence level greater than the pre-determined confidence level, the machine learning tool is further configured to:

prompt a user to select a comparable external dependency from the plurality of comparable external dependencies having a confidence level greater than the pre-determined confidence level;

assign the user-selected comparable external dependency a highest level of confidence;

store, in a database on the centralized server, the user-selected comparable external dependency and its associated external dependent metadata file; and

download the user-selected comparable external dependency.

5. The method of claim 1 wherein, when the application source code for an application is being compiled again for the user, the system further comprises automatically retrieving a stored user-selected comparable external dependency.

6. The method of claim 1 wherein, the application referenced external dependencies further include open-source software libraries, software objects and software classes.

7. A method for compiling an application, the method comprising:

retrieving from a centralized server;

an application source code associated with the application, the application source code referencing;

a plurality of internal dependencies stored in a cloud, the cloud in synchronization with the centralized server; and

a plurality of external dependencies for compiling the application;

an application metadata file including metadata associated with each of the pluralities of external dependencies; and

performing a plurality of machine learning steps using a machine learning tool, the plurality of steps comprising;

identifying the external dependencies referenced in the application;

instructing a web-crawler, the web-crawler being a component of the machine learning tool, to;

locate a plurality of comparable external dependencies for each of the identified referenced external dependencies, the comparable external dependencies provided by external frameworks on the internet; and

retrieve an external dependent metadata file for each of the plurality of located comparable external dependencies, the external dependent metadata file including metadata associated with the comparable external dependencies;

comparing the metadata of each comparable external dependent metadata file to the metadata of the application metadata file;

assigning, based on the comparison, a confidence level for each comparable external dependency, the confidence level being relative to a pre-determined confidence level, the assigning further comprising, assigning a greater confidence level to one of the located comparable external dependencies having a smallest file size relative to the other located comparable external dependencies;

downloading the located comparable external dependencies for each referenced external dependency having a confidence level greater than the pre-determined confidence level; and

combining and compiling the application source code with the internal dependencies and the downloaded external dependencies.

8. The method of claim 7 wherein, for each referenced external dependency in the application metadata file and for each of the external dependent metadata files, the metadata includes:

a file name;

a file size; and

a file version.

9. The method of claim 7 wherein the comparing of the metadata further comprises, comparing a file name, file size and file version of the referenced external dependencies to a file name, file size and file version of the external dependent metadata files for determining the confidence level.

10. The method of claim 7 wherein, when there are a plurality of comparable external dependencies being assigned a confidence level greater than the pre-determined confidence level, the machine learning tool is further configured to:

prompt a user to select a comparable external dependency from the plurality of comparable external dependencies having a confidence level greater than the pre-determined confidence level;

assign the user-selected comparable external dependency a highest level of confidence;

store, in a database on the centralized server, the user-selected comparable external dependency and its associated external dependent metadata file; and

download the user-selected comparable external dependency.

11. The method of claim 7 wherein, when the application source code for an application is being compiled again for the user, the system further comprises automatically retrieving a stored user-selected comparable external dependency.

12. The method of claim 7 wherein, the application referenced external dependencies further include open-source software libraries, software objects and software classes.

13. A method for compiling an application, the method comprising:

retrieving from a centralized server;

an application source code associated with the application, the application source code referencing;

a plurality of internal dependencies stored in a cloud, the cloud in synchronization with the centralized server; and

a plurality of external dependencies for compiling the application;

an application metadata file including metadata associated with each of the pluralities of external dependencies; and

performing a plurality of machine learning steps using a machine learning tool, the plurality of steps comprising;

identifying the external dependencies referenced in the application;

instructing a web-crawler, the web-crawler being a component of the machine learning tool, to;

locate a plurality of comparable external dependencies for each of the identified referenced external dependencies, the comparable external dependencies provided by external frameworks on the internet; and

retrieve an external dependent metadata file for each of the plurality of located comparable external dependencies, the external dependent metadata file including metadata associated with the comparable external dependencies;

comparing the metadata of each comparable external dependent metadata file to the metadata of the application metadata file;

assigning, based on the comparison, a confidence level for each comparable external dependency, the confidence level being relative to a pre-determined confidence level, the assigning further comprising, assigning a greater confidence level to one of the located comparable external dependencies having a latest version relative to the other located comparable external dependencies;

downloading the located comparable external dependencies for each referenced external dependency having a confidence level greater than the pre-determined confidence level; and

combining and compiling the application source code with the internal dependencies and the downloaded external dependencies.

14. The method of claim 13 wherein, for each referenced external dependency in the application metadata file and for each of the external dependent metadata files, the metadata includes:

a file name;

a file size; and

a file version.

15. The method of claim 13 wherein the comparing of the metadata further comprises, comparing a file name, file size and file version of the referenced external dependencies to a file name, file size and file version of the external dependent metadata files for determining the confidence level.

16. The method of claim 13 wherein, when there are a plurality of comparable external dependencies being assigned a confidence level greater than the pre-determined confidence level, the machine learning tool is further configured to:

prompt a user to select a comparable external dependency from the plurality of comparable external dependencies having a confidence level greater than the pre-determined confidence level;

assign the user-selected comparable external dependency a highest level of confidence;

store, in a database on the centralized server, the user-selected comparable external dependency and its associated external dependent metadata file; and

download the user-selected comparable external dependency.

17. The method of claim 13 wherein, when the application source code for an application is being compiled again for the user, the system further comprises automatically retrieving a stored user-selected comparable external dependency.

18. The method of claim 13 wherein, the application referenced external dependencies further include open-source software libraries, software objects and software classes.