Am Freitag, den 22.04.2005, 11:46 +0530 schrieb Avik Sengupta:
And we've seen that happen to POI in the last year that we've had
breakage in HEAD.. it really affects development velocity when you cant
answer the above question confidently!
I've tried to find a middle path in the current
Instead someone should fix the bugs.
Well, yes, that is ideal.. but if that doesnt happen, then you are left
with the possibility that regressions get into the codebase when people
cant answer Do all tests that are SUPPOSED TO PASS, continue to pass
after my fix?
And we've seen that happen to
with the following error. Ideas?? Can someone confirm this with latest
CVS HEAD?
(ant -Dtestcase=org.apache.poi.poifs.filesystem.TestEmptyDocument
single-test)
[junit] Testcase: testSingleEmptyDocument took 0.113 sec
[junit] Caused an ERROR
[junit] Cannot remove block[ 0 ]; out of
Ok, thanks. That makes sense. Do you have a bug he for this? I'll update the
code with that. In any event i think we should not run known failing tests by
default, it can hide other breakage. I've been thinking of a way to keep them
separately in cvs; for now i'll disable them.Quoting Rainer
Ok, thanks. That makes sense. Do you have a bug he for this? I'll update the
code with that. In any event i think we should not run known failing tests by
default, it can hide other breakage. I've been thinking of a way to keep them
separately in cvs; for now i'll disable them.Quoting Rainer
Ok, thanks. That makes sense. Do you have a bug he for this? I'll update the
code with that. In any event i think we should not run known failing tests by
default, it can hide other breakage. I've been thinking of a way to keep them
separately in cvs; for now i'll disable them.Quoting Rainer
Am Mittwoch, den 20.04.2005, 23:17 +0530 schrieb
[EMAIL PROTECTED]:
Ok, thanks. That makes sense. Do you have a bug he for this? I'll update the
code with that. In any event i think we should not run known failing tests by
default, it can hide other breakage. I've been thinking of a way to
