fix issues #316, #317.

[potterzot]

Description

This addresses issue #316 and also makes a quick fix to #317. The primary issue was that the longitude coordinates were not correctly being handled in the conversion from the -180 - 180 coordinate system to the 0 - 360 system. The program was taking the max() when it should have been taking the min(). However, on review there was a much simpler way of handling the indexing that more closely linked directly to values. The result is both easier to read and less error prone.

This PR also includes a new test to check the latitude and longitude values are those queried.

[sckott]

thanks @potterzot - LGTM
fix #316
fix #317

[sckott]

@potterzot i missed that the tests don't actually pass, did they pass for you or did you not run them on your end?

test-gefs.R:30: warning: gefs latitude and longitude selection returns correct values
longer object length is not a multiple of shorter object length

test-gefs.R:30: failure: gefs latitude and longitude selection returns correct values
all(unique(a$data$lat) == c(54, 54, 54, 54)) isn't true.

test-gefs.R:44: failure: gefs time and ensemble selection returns correct indices.
unique(d$data$ens) not equal to ens_idx - 1.
Attributes: < Modes: list, NULL >
Attributes: < Lengths: 1, 0 >
Attributes: < names for target but not for current >
Attributes: < current is not list-like >

test-gefs.R:58: failure: gefs metadata
d$dimensions[1:4] not equal to c("lon", "lat", "height_above_ground", "ens").
target is NULL, current is character

Apologies I should have looked more closely at this PR. Can you send a new PR? Some suggestions:

• For the test on line 30 i think you want %in% instead of ==. Also the test doesn't pass so not sure if it should be changed or what?
• For the test on line 44, the left side results in an integer and the right side a numeric, so change the test to account for that. Possibly you changed something in this PR that changed the test result
• For the test on line 58, the dimensions slot is NULL. i assume that's related to changes in this PR?

[potterzot]

Ack sorry, they passed on my end but I must have done something after that but before the commit. Sorry! I'll look into it asap.