New Method visualize.contacts and extensive testing

[gph82]

[marscher]

now we have "random" hangs at the notebook tests...

[gph82]

yes this is the interval_schachtelung, I suspect. I'll factor in some condition to exit when the schachtelung gets trapped

[codecov-io]

Codecov Report

Merging #47 into master will increase coverage by 0.12%. The diff coverage is 94.14%.

@@            Coverage Diff             @@
##           master      #47      +/-   ##
==========================================
+ Coverage    96.1%   96.22%   +0.12%     
==========================================
  Files          13       13              
  Lines        2129     2491     +362     
==========================================
+ Hits         2046     2397     +351     
- Misses         83       94      +11

Impacted Files	Coverage Δ
molpx/generate.py	100% <100%> (ø)	:arrow_up:
molpx/tests/test_visualize.py	99.46% <100%> (+0.06%)	:arrow_up:
molpx/tests/test_bmutils.py	99.82% <100%> (+1.6%)	:arrow_up:
molpx/_bmutils.py	99.58% <100%> (+3.57%)	:arrow_up:
molpx/tests/test_linkutils.py	87.02% <73.59%> (-11.22%)	:arrow_down:
molpx/tests/test_generate.py	96% <96%> (ø)
molpx/visualize.py	99.4% <97.36%> (+0.05%)	:arrow_up:
molpx/_linkutils.py	97.98% <98.43%> (+0.11%)	:arrow_up:
... and 1 more

---

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 df39f89...7f30a3a. Read the comment docs.

[gph82]

The tests do get hang up (e.g. here for 15 min https://travis-ci.org/markovmodel/molPX/jobs/383598850#L970) but that line runs quickly on other tests and locally in my machine. I'd merge this for now (new features are coming) and see if we get more info on the way to the next release (soon)

[marscher]

Develop new stuff without functional CI will cause headaches later on (just my honest opinion).

[gph82]

So how do we find out where/why it is hanging?

[marscher]

execute with --boxed and --timeout=120 (the latter would require pytest-timeout or so)

[gph82]

you suggest we do that locally first, right?

[marscher]

On 05/25/2018 12:12 PM, Guillermo Pérez-Hernández wrote:

you suggest we do that locally first, right? I thought the test are only hanging remotely?

[gph82]

At some weird, irreproducible moments, they also hang locally