Show node that originated a warning in the warnings summary

[nicoddemus]

Weird that nobody has reported this regression until now (and embarrassing that I did not notice this myself earlier!)

Here's how it looks on master:

========================= warnings summary ==========================
c:\pytest\.tmp\test-warning-loc.py:5: UserWarning: This is a test warning
  warnings.warn("This is a test warning")

And with this PR:

========================= warnings summary ==========================
test-warning-loc.py::test_foo
  c:\pytest\.tmp\test-warning-loc.py:5: UserWarning: This is a test warning
    warnings.warn("This is a test warning")

Fix #4221

[blueyed]

Flake complains:

src/_pytest/debugging.py:4:1: F401 'os' imported but unused

Not related to this PR though apparently.

[nicoddemus]

Oh bummer. Fixing in sec...

[asottile]

heh, an implicit merge conflict with the debugger branch 🙃

[nicoddemus]

Oh bummer. Fixing in sec...

Or I will rebase on #4231. 😁

[nicoddemus]

Done, should be fine now. 👍

[codecov]

Codecov Report

❗ No coverage uploaded for pull request base (master@15278aa). Click here to learn what that means.
The diff coverage is 100%.

@@           Coverage Diff            @@
##             master   #4229   +/-   ##
========================================
  Coverage          ?   93.6%           
========================================
  Files             ?     109           
  Lines             ?   24537           
  Branches          ?    2412           
========================================
  Hits              ?   22969           
  Misses            ?    1245           
  Partials          ?     323

Flag	Coverage Δ
#linux	93.47% <100%> (?)
#py27	91.53% <100%> (?)
#py34	92.12% <100%> (?)
#py36	92.11% <100%> (?)
#windows	91.04% <100%> (?)

Impacted Files	Coverage Δ
testing/test_warnings.py	98.55% <ø> (ø)
src/_pytest/terminal.py	91.08% <100%> (ø)

---

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 15278aa...2ad43ee. Read the comment docs.

[RonnyPfannschmidt]

seems to be a travis timeout

[RonnyPfannschmidt]

it seems the appveyor failure is neglect-able here