what can we do when we
play

WHAT CAN WE DO WHEN WE LACK ADEQUATE TESTS ? 58 th CREST Open - PowerPoint PPT Presentation

Aug 17, 2022 •440 likes •513 views

WHAT CAN WE DO WHEN WE LACK ADEQUATE TESTS ? 58 th CREST Open Workshop Breakout discussion summary CREATE TESTS EvoSuite Record real world usage Re-use tests from tests across (compatible) components Metamorphic testing

  1. WHAT CAN WE DO WHEN WE LACK ADEQUATE TESTS ? 58 th CREST Open Workshop Breakout discussion summary

  2. CREATE TESTS • EvoSuite • Record real world usage • Re-use tests from tests across (compatible) components • Metamorphic testing • Symbolic execution • Isomorphic Regression T esting

  3. MEASURE TESTS See COW 57 on Information Theory and testing to measure the quality of tests and how to create diversity in test suites.

  4. BOTS Watson Watcher: is bot that recognizes bad patches, bad because they are unlikely to solve the issue or because they might “consume” a lot of code review time.

  5. BOTS Watson Watcher: is bot that recognizes bad patches, bad because they are unlikely to solve the issue or because they might “consume” a lot of code review time. Patrick Patcher: is a bot that creates patches (APR) for bugs and compare them to the human produced patch to improve quality over time. Sits on the background for a long time and at some point can become active in the code review process.

  6. BOTS Watson Watcher: is bot that recognizes bad patches, bad because they are unlikely to solve the issue or because they might “consume” a lot of code review time. Patrick Patcher: is a bot that creates patches (APR) for bugs and compare them to the human produced patch to improve quality over time. Sits on the background for a long time and at some point can become active in the code review process. Galois: is a bot that can create a high level specification and that the programmer/tester can ask questions in “English”.

Recommend

More recommend