Title :
Extracting Code Examples from Unit Test Cases
fDate :
Sept. 29 2014-Oct. 3 2014
Abstract :
Understanding how to properly use APIs of large libraries is difficult, error prone, and time consuming. Software developers resort to study to learn APIs. Several approaches have been proposed to mine these examples, but the sources from which they mine examples as well as their mining approaches hamper their applicability in some practical scenarios. Unit test cases seem to be an additional source of significant API examples, which may overcome the aforementioned difficulties. Synthesizing meaningful examples from tests not only improves the applicability of current code recommendation systems, but also facilitates providing up to date API examples to augment documentation. However, mining examples of API use from unit tests is a non trivial task and arises several research challenges summarized in this paper.
Keywords :
application program interfaces; data mining; program testing; source code (software); API; code example extraction; current code recommendation systems; meaningful example synthesis; mining approach; nontrivial task; unit test cases; Conferences; Data mining; Documentation; Libraries; Software maintenance; Testing; code examples; code recommendation; unit test cases;
Conference_Titel :
Software Maintenance and Evolution (ICSME), 2014 IEEE International Conference on
Conference_Location :
Victoria, BC
DOI :
10.1109/ICSME.2014.122