Add a parameter to run the agent on proxy creation

[ocaballeror]

I'm not sure I understood issue #153 correctly, but this is what I came up with

Edit: fixes #153.

[ocaballeror]

I am aware that some tests fail and that the whole test suite seems to run into some kind of deadlock, but this is just the first attempt at this. I'll update the branch when I know this is the right way to do things.

[Peque]

It seems Travis is down... :confused:

Yeah, I think you misunderstood the issue. When in doubt... just ask! :wink: It is only normal to misunderstand some issues/conversations if you were not there the day they started. It is also probably our fault for not being clearer in the description.

Updated the issue to, hopefully, make it more clear.

Also, all changes should have tests validating the new functionality, so have a look at the test suite and create a test that checks what you want to implement. You might want to create the tests first, let me review them, and then implement the fix that makes the new tests pass. :blush:

[codecov[bot]]

Codecov Report

Merging #221 into master will increase coverage by <.01%. The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #221      +/-   ##
==========================================
+ Coverage   99.01%   99.02%   +<.01%     
==========================================
  Files          25       25              
  Lines        3356     3388      +32     
  Branches      244      245       +1     
==========================================
+ Hits         3323     3355      +32     
  Misses         22       22              
  Partials       11       11

Impacted Files	Coverage Δ
osbrain/tests/test_proxy.py	100% <100%> (ø)	:arrow_up:
osbrain/agent.py	97.5% <100%> (-0.01%)	:arrow_down:
osbrain/proxy.py	100% <100%> (ø)	:arrow_up:

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update c622f65...ddc3bbb. Read the comment docs.

[ocaballeror]

I reviewed the tests. The final distribution is:

• A parametrized test for wait_for_running with 1 and -1
• A separate one that tests 0 as timeout
• A third test that triggers a timeout, parametrized with 0 and 1. It didn't make sense to move 0 to a separate test here.