Failure with dev waldo

[hadley]

I see

Failure (test-getLambda.R:26:3): lambda calcs. are correct
lambdas$absolute_basalArea_tplus1/lambdas$absolute_basalArea_t (`actual`) not equal to lambdas$lambda (`expected`).

  `actual[13:19]`: 1 7 2 NaN Inf 0 Inf
`expected[13:19]`: 1 7 2  NA Inf 0 Inf

Because waldo now correctly distinguishes between NA_real_ and NaN. I took a look at the test and it doesn't look like there's an obvious fix there since you're using precomputed reference data. This may suggest that there's a genuine (small) bug in your code.

[aestears]

Fixed in the current dev branch: The getLambda() function previously changed NaNs to NAs, but that code has been removed... both to pass the test completely, but also to differentiate from true NAs (i.e. one of the values in t or t+1 are NAs) and NaNs that occur because the t+1 value is 0 https://github.com/aestears/plantTracker/commit/d9ac9da30d77af4cf70923fa808f4956f9cfc8fd