[Twisted-Python] Error (and response) handling in protocols

[Michael Thompson]

On 21 February 2011 14:06, Jason Heeris <jason.heeris at gmail.com> wrote:
> On 21 February 2011 21:30,  <exarkun at twistedmatrix.com> wrote:
>> Not intentionally:
>>
>>  http://twistedmatrix.com/trac/ticket/3690
>
> Interesting...
>
> As an aside, the problem of serial port availability for unit tests is
> something I've encountered before too. I solved it by using "socat" to
> create virtual serial ports with known names (paths on *nix, COM
> numbers on Windows), and ran my tests using that. In fact, the test
> for the patch in #3690 only requires that the port exists, not that it
> does anything, which is even easier than actually requiring the port
> to do anything.

I'm in the process of writing a mock that will allow the serial port
testing to work without a physical port as I think this is the best
solution for the Twisted build farm. I have the posix part done and a
plan for the windows version, using named pipes.