Loading...
Please wait, while we are loading the content...
Conference abstracts
| Content Provider | ACM Digital Library |
|---|---|
| Editor | Slamecka, Vladimir |
| Abstract | One problem in computer program testing arises when errors are found and corrected after a portion of the tests have run properly. How can it be shown that a fix to one area of the code does not adversely affect the execution of another area? What is needed is a quantitative method for assuring that new program modifications do not introduce new errors into the code. This model considers the retest philosophy that every program instruction that could possibly be reached and tested from the modified code be retested at least once. The problem is how to determine the minimum number of test cases to be rerun. The process first involves generating the test case dependency matrix and the reachability matrix. Using the test case dependency matrix and the appropriate rows of the reachability matrix, a 0-1 integer program can be specified. The solution of the integer program yields the minimum number of test cases to be rerun, and the coefficients of the objective function identify which specific test cases to rerun. |
| Starting Page | 1 |
| Ending Page | 36 |
| Page Count | 36 |
| File Format | |
| DOI | 10.1145/800008.808038 |
| Language | English |
| Publisher | Association for Computing Machinery (ACM) |
| Publisher Date | 1977-01-01 |
| Publisher Place | New York |
| Access Restriction | Subscribed |
| Content Type | Text |