Missing 0.10 docs

[svenevs]

Continuation of #13 (I never documented the changes...and completely forgot sorry!).

• [X] Fixes #16 updated parametrize test case.
  • [X] Update docs on parametrize for how to do it in the new world order.
• [X] Document the new setup_raises (especially limitations).
• [X] Document the new match argument for raises and setup_raises.

@JamesMakela-NOAA @AetherUnbound regarding #16 this commit is what you are talking about right? AKA parametrize can still be done, we just need to use the new way in the tests (and change the README documentation) right?

[codecov[bot]]

Codecov Report

Merging #17 into master will increase coverage by 0.04%. The diff coverage is 100%.

@@            Coverage Diff             @@
##           master      #17      +/-   ##
==========================================
+ Coverage   97.97%   98.01%   +0.04%     
==========================================
  Files           2        2              
  Lines          99      101       +2     
==========================================
+ Hits           97       99       +2     
  Misses          2        2

Impacted Files	Coverage Δ
tests/test_raises.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 ee34659...f1ca5a5. Read the comment docs.

[svenevs]

Ok this looks good to me now, but people should review the verbiage to make sure it is clear.

For convenience, you can see the rendered markdown on svenevs/pytest-raises@missing_0.10_docs branch. I moved a few paragraphs up to just under the main Usage section and then expanded from there, I think it's clear.

What I'm not sure about is the setup_raises example. It's extra contrived, but the thing is describing the real-world scenario in which I needed it is not helpful. It involves quite a bit of machinery, and knowledge of the sphinx build process...so I thought this was better than just a bare raise in pytest_runtest_setup but without making it super complicated.

Note that I'm installing latest pytest on travis now, otherwise the setup_raises sample code would have to include the get_closest_marker vs get_marker split code we have which would only make it more confusing. It also verifies that with pytest 4 the parameterize examples are indeed working now :slightly_smiling_face:

[Lemmons]

Looks great. Big improvement to the documentation. Thanks!