Set run arguments in .coveragerc

[nedbat]

<donri> nedbat: i'm the author of a test automation framework and it'd be nice to support coverage without doing anything special and still have usage be simple; thus i thought .coveragerc configured runner was a good idea
<donri> "add these two lines to .coveragerc and you can measure coverage with coverage run"
<nedbat> donri: i can see the appeal, and I don't think it would be tricky to do.  It'd just never occurred to me someone would want it.
<nedbat> donri:  it would also be useful for projects where you need to provide specific args when doing tests, or when doing coverage tests.
<donri> yep
<donri> in my own project i solved it with a paver task, which also does report and html in one go, so i'm satisfied on a personal level
<donri> still would be nice if it integrated elegantly with my test tool

(donri == Dag Odenhall)

---

• Bitbucket: https://bitbucket.org/ned/coveragepy/issue/134

[nedbat]

Original comment by Dag Odenhall (Bitbucket: dag, GitHub: dag)

---

I should add that in my particular case it is primarily the -m switch which is interesting. For reference there's a related ticket in my project: dag/attest#77

[nedbat]

Putting run arguments in the .coveragerc doesn't feel right to me.