search

BiologicalRecordsCentre / sparta

Species Presence/Absence R Trends Analyses
http://biologicalrecordscentre.github.io/sparta/index.html
MIT License
21 stars 24 forks source link

More tests failing #208

Closed drnickisaac closed 3 years ago

drnickisaac commented 3 years ago

Tests are failing on the newest version of R. This includes test failures in parts of the code that have not changed, e.g. testTelfer and dataDiagnostics. Also, in testformatOccData the tests comparing head(...) with head_ are failing but they look identical when printed to screen. I think this must be related to more stringent language settings in R v4.

drnickisaac commented 3 years ago

Back at home on my mac there are now just the previous four tests failing on testdetection_phenology.

AugustT commented 3 years ago

Look at str of the outputs being compared, are they different? Play with expectequal expectidentical and other variants as some are less strict that others. When testing at home make sure you are testing with the same version of R that Travis is.

drnickisaac commented 3 years ago

The Travis tests have passed, even for the ones that fail on my home machine (a Mac). Some AppVeyor tests fail on the development version of R. At work there were other tests failing, including on scripts that have not changes (e.g. Telfer) - one possibility is that these are related to the AppVeyor fails. Another possibility is that those test failures were downstream consequences of changes I was making at the time (unlikely but possible). I will look at the structures, as you suggest. If you were able to make sense of the AppVeyor fail then we can come to a view.

drnickisaac commented 3 years ago

Travis says Yes!