Test relevancy prediction for code changes
AssigneeEntlT Software LLC
InventorsGabi Shalev, Itay Ben-Yehuda, Raz Regev, Elad Benedict
Life of this patent
Estimating the term…
Abstract
In some examples, test relevancy prediction for code changes may include ascertaining files for a commit for a build, and for each test of a plurality of tests, determining a score based on a weight assigned to a file of the ascertained files. Test relevancy prediction for code changes may further include ordering each test of the plurality of tests according to the determined score, and identifying, based on the ordering of each test of the plurality of tests, tests from the plurality of tests for which the score exceeds a specified threshold. The identified tests may represent tests that are to be applied to the build.
Claims20
An apparatus comprising:
- a processor;
- and
- a non-transitory computer readable medium storing machine readable instructions that when executed by the processor cause the processor to:
- ascertain files for a commit for a build;
- for each test of a plurality of tests, determine a score based on a weight assigned to a file of the ascertained files;
- order each test of the plurality of tests according to the determined score;
- and
- identify, based on the ordering of each test of the plurality of tests, tests from the plurality of tests for which the score exceeds a specified threshold, wherein the identified tests represent tests that are to be applied to the build.
A method comprising:
- identifying, by a processor, a test and file pair in a set of builds, wherein the set of builds is associated with a failed test and a passed test;
- and
- for the identified test and file pair, logging an occurrence of a file included in the test and file pair, the logging associated with the failed test and the passed test.
A non-transitory computer readable medium having stored thereon machine readable instructions, the machine readable instructions, when executed, cause a processor to:
- ascertain files for each commit for a build;
- for each test of a plurality of tests, determine, based on a trained model that accounts for a failed test and passed test, a score based on a weight assigned to a file of the ascertained files;
- order each test of the plurality of tests according to the determined score;
- and
- identify, based on the ordering of each test of the plurality of tests, tests from the plurality of tests for which the score exceeds a specified threshold, wherein the identified tests represent tests that are to be applied to the build.
Dependencies are inferred from the claim text.