TST: Fix warnings

[adamrp]

The message of the warning could also be checked, if desired, but previously
it was not being checked explicitly.

[adamrp]

There are other places where npt.assert_warns is used, but so far none of those places is posing a problem (since the "desired" warning is the first one that is raised, apparently). Do we want to go ahead and replace all the calls to npt.assert_warns with a construct similar to what's here?

[ElDeveloper]

I'm not sure I like this approach, in general our code should not
produce warnings and if we are testing for a warning in specific, then
that's the only warning that should be raised in that test case.

On (May-14-15|16:07), adamrp wrote:

There are other places where npt.assert_warns is used, but so far none of those places is posing a problem (since the "desired" warning is the first one that is raised, apparently). Do we want to go ahead and replace all the calls to npt.assert_warns with a construct similar to what's here?

---

Reply to this email directly or view it on GitHub:
#1182 (comment)

[adamrp]

You're right, I'll try to figure out if it's something in our code that's
eliciting the futurewarning from pandas. But in the meantime, perhaps we
should consider this aproach as a stop gap so that our builds pass and we
can gain added confidence merging new code?
On May 14, 2015 7:04 PM, "Yoshiki Vázquez Baeza" notifications@github.com
wrote:

I'm not sure I like this approach, in general our code should not
produce warnings and if we are testing for a warning in specific, then
that's the only warning that should be raised in that test case.

On (May-14-15|16:07), adamrp wrote:

There are other places where npt.assert_warns is used, but so far none
of those places is posing a problem (since the "desired" warning is the
first one that is raised, apparently). Do we want to go ahead and replace
all the calls to npt.assert_warns with a construct similar to what's here?

---

Reply to this email directly or view it on GitHub:
#1182 (comment)

—
Reply to this email directly or view it on GitHub
#1182 (comment).