On 08/05/2019 10:52 pm, Andrew Stitcher wrote:
On Wed, 2019-05-08 at 10:41 -0400, Andrew Stitcher wrote:
On Wed, 2019-05-08 at 13:29 +0100, Gordon Sim wrote:
...
* ran all python examples (against qpid-cpp broker for those that
    require a broker)
    - the helloworld_direct_tornado.py no longer works (it does work
with
      0.27.1)
      Not sure whether this is deliberate or expected.

It is expected to work. It certainly was working when the python
changes were initially committed. I will give this a try and see
what's
going on.

The work to fix the reconnect behaviour (and reinstate non blocking
connect) did indeed break the tornado examples.

I would make this release blocking except that the tornado examples
were actually broken for a long time (afaict) before I fixed them in
SHA 603ce7a7 (PROTON-1991) which was only 4 months ago so I'm not
actually sure how many people really care about them.

On the other hand it is a regression from 0.27, but not exactly a
regression for some earlier releases (but with different problems).

I've gone back through 0.27.1, 0.26.0, 0.25.0, 0.24.0 and 0.23.0 and in all those helloworld_tornado.py and helloworld_direct_tornado.py work though you are right that the recurring_timer_tornado.py example did not.

The test_examples run helloworld_tornado.py and helloworld_direct_tornado.py and I run that pretty regularly (certainly most proton releases I vote on) so I would have been surprised if they had been broken. (The recurring timer example is not automated and I never really run that one).

Gordon do you want to make this failure -1 the release and respin when
I have a fix?

Whats the nature of the issue/fix? Is it only these examples that would be affected?

My inclination would be to respin with the fix. Though I would readily concede that the examples are corner cases, I believe they have been working through recent releases (apart from the timer one) and it would be a shame to break that without some reason.

---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to