US 11,455,566 B2Granted

Classifying code as introducing a bug or not introducing a bug to train a bug detection algorithm

AssigneeInternational Business Machines Corporation

InventorsAlexander Sobran, Bo Zhang

Life of this patent

Estimating the term…

Loading patent…

Abstract

Provided are a computer program product, system, and method for classifying code as introducing a bug or not introducing a bug to train a bug detection algorithm. For each commit in a commit history of code changes to the code base, a determination is made of lines of code changed by the commit. For each line of code changed by the commit, a determination is made as to whether the commit is for a bug fix. A determination is made as to whether a previous commit changing the line of code changed by the commit for the bug fix in response to determining that the commit is for the bug fix. Indication is made that the previous commit introduced a bug. The algorithm is trained to classify changes to lines of code by commits indicated as having introduced a bug as bug introducing commits.

Claims23

Dependent claims

A computer program product for training an algorithm to determine whether code changes to a code file will result in a code error, the computer program product comprising a computer readable storage medium having computer readable program code embodied therein that is executable to perform operations, the operations comprising:

providing a line change list having entries for lines of code changed by commits, wherein an entry in the line change list identifies a line of code and a commit that changes the line of code, wherein commits change lines of code in the code file;
for a line of code changed by a commit comprising a bug fix, perform:
adding an entry in the line change list indicating the line of code and the commit comprising the bug fix that changes the line of code;
processing the line change list to determine a previous entry to the added entry in the line change list indicating the line of code in the added entry and a previous commit making a previous change to the line of code indicated in the added entry;
indicating in the entry in the line change list for the previous commit and the line of code that the previous commit introduced a bug;
indicating the previous commit as introducing a bug;
and
indicating that the code file, including the line of code changed by the previous commit, as introducing a bug;
and
training the algorithm to perform at least two classifications that are a member of a set of classifications comprising:
classify code files, indicated as introducing a bug, as introducing a bug;
classify commits, indicated as introducing a bug, as introducing a bug;
and
classify lines of code, indicated in entries in the line change list as introducing a bug, as introducing a bug.

11 more elements

A system for training an algorithm to determine whether code changes to a code file will result in a code error, comprising:

a processor;
and
a computer readable storage medium having computer readable program code that when executed by the processor performs operations, the operations comprising:
providing a line change list having entries for lines of code changed by commits, wherein each entry in the line change list identifies a commit and a line of code changed by the identified commit, wherein commits change lines of code in the code file;
for a changed line of code indicated in the change line list that is changed by a commit comprising a bug fix, perform:
processing the line change list to determine a previous entry for the changed line of code indicating a previous commit making a previous change to the changed line of code;
indicating in the entry in the line change list for the previous commit and the changed line of code that the previous commit introduced a bug;
indicating the previous commit as introducing a bug;
and
indicating that the code file, including the line of code changed by the previous commit, as introducing a bug;
and
training the algorithm to perform at least two classifications that are a member of a set of classifications comprising:
classify code files, indicated as introducing a bug, as introducing a bug;
classify commits, indicated as introducing a bug, as introducing a bug;
and
classify lines of code, indicated in entries in the line change list as introducing a bug, as introducing a bug.

13 more elements

A method for training an algorithm executed in a computer system to determine whether code changes to a code file will result in a code error, comprising:

providing a line change list having entries for lines of code changed by commits, wherein each entry in the line change list identifies a commit and a line of code changed by the identified commit, wherein commits change lines of code in the code file;
for a changed line of code indicated in the change line list that is changed by a commit comprising a bug fix, perform:
processing the line change list to determine a previous entry for the changed line of code indicating a previous commit making a previous change to the changed line of code;
indicating in the entry in the line change list for the previous commit and the changed line of code that the previous commit introduced a bug;
indicating the previous commit as introducing a bug;
and
indicating that the code file, including the line of code changed by the previous commit, as introducing a bug;
and
training the algorithm to perform at least two classifications that are a member of a set of classifications comprising:
classify code files, indicated as introducing a bug, as introducing a bug;
classify commits, indicated as introducing a bug, as introducing a bug;
and
classify lines of code, indicated in entries in the line change list as introducing a bug, as introducing a bug.

10 more elements

Dependencies are inferred from the claim text.