On Mon, Feb 8, 2010 at 23:48, Richard Holland <[email protected]> wrote:
(taverna-hackers material - moved from taverna-users) > Would it be worth maintaining a manual test suite for testing GUI bugs? When > a GUI bug is reported, the manual test suite would gain a new test explaining > how to reproduce the bug, and what it looks like, and what the correct > behaviour should be. Then before each new release the manual test suite would > be rerun (by hand, obviously) alongside all the usual JUnits to make sure > there is no regression. > > Probably a bit time consuming though, but worth it to improve software > quality? In theory, but the problem is that just going through such a list would take several weeks. Just testing the GUI fixes for a certain release takes several days.. That list could just be every bug from T2-1 and upwards.. that's almost 1100 issues just there, so some kind of automated GUI testing would be needed, but I've not had a chance to try the various suites who claim to be able to do scripted swing component testing. That said, it should still be possible to unit test most back-end features without invoking the GUI if you write your code structured enough. For instance, let's imagine you are fixing a bug with renaming processors, where for some reason Taverna allowed you to create two processors called "fish1" and "Fish1", and you want the duplication detection to be case insensitive. If you have a separate function or class that does the name checking, it's easy to write a unit test that tries to create both "fish1" and "Fish1" - but if that check is done in the middle of the click-action it's very hard to test without constructing bits of the GUI. (which in many cases can lead to setting up the whole workbench) -- Stian Soiland-Reyes, myGrid team School of Computer Science The University of Manchester ------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ taverna-hackers mailing list [email protected] Web site: http://www.taverna.org.uk Mailing lists: http://www.taverna.org.uk/taverna-mailing-lists/ Developers Guide: http://www.mygrid.org.uk/tools/developer-information
