terrycojones/gist:2177560

## gistfile1.weechatlog
[2012-03-24 01:28] <terrycojones> exarkun: thanks for fielding the question in stackoverflow related to txRDQ.  i'm trying to figure out what's going wrong.
[2012-03-24 01:30] <terrycojones> exarkun: i've been putting in print stmts for about 30 mins and everything seems fine, but trial for some reason is printing [ERROR].
[2012-03-24 01:32] <idnar> terrycojones: the usual cause of an [ERROR] when there's apparently no exception being raised is that an exception was *logged* and not flushed
[2012-03-24 01:32] <idnar> (at least, in my experience)
[2012-03-24 01:32] <idnar> (I have absolutely no context here though, so that's just a wild guess)
[2012-03-24 01:33] <terrycojones> idnar: that sounds plausible / in line with exarkun's ticket suggestion, thanks. you just caused me to go read what exarkun wrote a bit more closely :-)
[2012-03-24 01:54] <terrycojones> idnar: well, i still don't see what the error is.  idnar context is at http://stackoverflow.com/questions/9728781/testing-a-failing-job-in-resizabledispatchqueue-with-trial if you feel like looking.
[2012-03-24 01:55] <terrycojones> idnar: the self.queue.put(...) call returns a deferred. if i addBoth to it with something that flushes logged errors, there are no logged errors.  even though the deferred has fired & failed.
[2012-03-24 01:56] <terrycojones> idnar: and i'm returning the deferred from the test case, to make sure trial gets it.
[2012-03-24 01:59] <terrycojones> idnar: i guess what i'm seeing would make sense if trial gave an [ERROR] after the test returns even if the test returns a deferred.  so it gets the test result back, complains about the logged error, and sometime later the deferred fires & the flush finds no errors because trial has complained and cleared the logged errors.
	[2012-03-24 01:28] <terrycojones> exarkun: thanks for fielding the question in stackoverflow related to txRDQ. i'm trying to figure out what's going wrong.
	[2012-03-24 01:30] <terrycojones> exarkun: i've been putting in print stmts for about 30 mins and everything seems fine, but trial for some reason is printing [ERROR].
	[2012-03-24 01:32] <idnar> terrycojones: the usual cause of an [ERROR] when there's apparently no exception being raised is that an exception was logged and not flushed
	[2012-03-24 01:32] <idnar> (at least, in my experience)
	[2012-03-24 01:32] <idnar> (I have absolutely no context here though, so that's just a wild guess)
	[2012-03-24 01:33] <terrycojones> idnar: that sounds plausible / in line with exarkun's ticket suggestion, thanks. you just caused me to go read what exarkun wrote a bit more closely :-)
	[2012-03-24 01:54] <terrycojones> idnar: well, i still don't see what the error is. idnar context is at http://stackoverflow.com/questions/9728781/testing-a-failing-job-in-resizabledispatchqueue-with-trial if you feel like looking.
	[2012-03-24 01:55] <terrycojones> idnar: the self.queue.put(...) call returns a deferred. if i addBoth to it with something that flushes logged errors, there are no logged errors. even though the deferred has fired & failed.
	[2012-03-24 01:56] <terrycojones> idnar: and i'm returning the deferred from the test case, to make sure trial gets it.
	[2012-03-24 01:59] <terrycojones> idnar: i guess what i'm seeing would make sense if trial gave an [ERROR] after the test returns even if the test returns a deferred. so it gets the test result back, complains about the logged error, and sometime later the deferred fires & the flush finds no errors because trial has complained and cleared the logged errors.