Victor,
I think the design should be a bit different.
Instead of passing 'exc_info', we can pass 'traceback' key, like
in the snippet below:
self.call_exception_handler({
'message': 'socket.accept() out of system resource',
'exception': exc,
'traceback': sys.exc_info()[2],
'socket': sock,
})
This way, apps written for Trollius will be easier to port
to asyncio (in 3.4.1 we can fix call_exception_handler to remove
traceback from the context, if ``context['traceback']`` is the same
as ``context['exception'].__traceback__``)
Yury
On 2/25/2014, 9:23 AM, Victor Stinner wrote:
Hi,
In Python 2, an exception has no __traceback__ attribute. Calling
call_exception_handler() with the exception is not enough to get the
traceback.
I propose to add new exc_info key to the context in Trollius. If
exc_info is True, pass sys.exc_info() to the logger. Could you please
review:
http://codereview.appspot.com/68560043
By the way, it looks like there is no unit test checking that the
exception handler logs a traceback. See also my issue proposing to log
where a Future was created, it adds some tests on the traceback (but
not directly related to call_exception_handler):
http://code.google.com/p/tulip/issues/detail?id=137
https://codereview.appspot.com/64900043/
Victor
