I can't see what's caused this, but a bunch of tests are failing because the JTPepub objects are "randomly" nil.
I think they're nil because mimetype (inside openEPUBfile) is wrong - NSlogging of mimetype shows:
Mar 2 21:43:03 aluminium otest[23787]: mimetype application/epub+zip
Mar 2 21:43:03 aluminium otest[23787]: mimetype application/epub+zip2007/ops
Mar 2 21:43:03 aluminium otest[23787]: mimetype application/epub+zipctions
Mar 2 21:43:03 aluminium otest[23787]: mimetype (null)
I tried using NULL instead of my mmapped I/O stuff, no change.
Something's up, I think with memory somewhere. I think the init methods for ZipArchive are screwy - self = [self init] for example. Deleting the (useless) init method might help. Also you must call [self release] on failures, not [self dealloc]. The unit tests call setUp and tearDown for every test case, so they do exercise memory quite a bit.
I'm not sure what the best way of debugging memory problems like this is in Xcode...
I can't see what's caused this, but a bunch of tests are failing because the JTPepub objects are "randomly" nil.
I think they're nil because mimetype (inside openEPUBfile) is wrong - NSlogging of mimetype shows:
Mar 2 21:43:03 aluminium otest[23787]: mimetype application/epub+zip Mar 2 21:43:03 aluminium otest[23787]: mimetype application/epub+zip2007/ops Mar 2 21:43:03 aluminium otest[23787]: mimetype application/epub+zipctions Mar 2 21:43:03 aluminium otest[23787]: mimetype (null)
I tried using NULL instead of my mmapped I/O stuff, no change.
Something's up, I think with memory somewhere. I think the init methods for ZipArchive are screwy - self = [self init] for example. Deleting the (useless) init method might help. Also you must call [self release] on failures, not [self dealloc]. The unit tests call setUp and tearDown for every test case, so they do exercise memory quite a bit.
I'm not sure what the best way of debugging memory problems like this is in Xcode...