| Issue 122: | ANT task does not handle syntax errors well | |
| 1 person starred this issue and may be notified of changes. | Back to list |
from bendavies@he3.com.au: Just a question to make sure I have my environment configured correctly and understand what I should expect the behaviour of the MXunit ant task to be. I notice that when my TestCase or my Tested Code has a syntax error (ie. Very frequently) the output from the Ant Task is not very useful - I need to switch to a browser and manually run the test to determine the actual cause of the problem. Because this interrupts my TDD rhythm (embryonic as it might be) and because syntax errors are common, I wonder if this behaviour is normal. Can the Ant task return information on Syntax errors in the TestCase or Errors thrown by the TestCase due to syntax or runtime errors in the tested code? ( I suspect what I am really asking is if the error handling of the RemoteFacade call should catch syntax errors in the TestCase or the Setup / Teardown methods and report them as errors or failures in the normal, controlled way). I hope that makes sense. I can explain more if required. If this currently is not supported by mxunit, and no one else feels like contributing, I may be able to modify the core files to effect this change, but I would like some guidance before I do ! Marc's note: this is all part of the clusterf*ck i created in DirectoryTestSuite by keeping "catastrophic" errors separate from all other problems. Bill, maybe the thing to do is inside of DirectoryTestSuite, after all tests run, if there are catastrophic errors like this one then we should merge them back into the primary error structure. sound like a plan?
Jun 30, 2008
Project Member
#1
marc.es...@gmail.com
Status:
Fixed
|