Create regression tests to allow changing your code confidently

[maurolepore]

From @maurolepore on September 18, 2018 15:35

One way to improve your functions with confidence that you are not breaking anything is to run regression tests -- tests that automatically compare the result of a computation against a know result. You can do this with testthat::expect_known_output(). It is a fairly advanced topic and I'm happy to help you.

Copied from original issue: ErvanCH/AGBflux_pack#16

[maurolepore]

@ervanSTRI, please run data_preparation() with the data in data/ and send me the output. That output will be my reference to ensure I don't change the behavior of your function when I do some change. So far I have been very conservative. I have not made any changes that could affect the output (only changed a bit the style of the code and added some comments). But to really help you I will need to make bigger changes. And I can't do that safely until I have a regression test.

I need your help with this because right now, I fail to fun your function. And I don't want to try fix it because in doing so I may break something else.

[maurolepore]

@ervanSTRI, I see that you added the folder output/ with some plots and a .csv file. The .csv file contains only what seems to be column-headers. Is that what you expect?

Also, I see the file "barro colorado island_formated_data.Rdata" in data/. When I load it its name is DF. Is this the output of data_preparaiton()?

As you advance in this and other issues, please leave a record here so I can quickly understand what is the state of each issue.

[maurolepore]

@ervanSTRI, is the object DF (named "barro colorado island_formated_data.Rdata" in data/. ) the one you are refering to when you write data(df): For example:

# data(df)
# correctDBH(df[treeID==240])

If so, notice that one is lowercase and the other one is uppercase. Also, I'm not clear about what DF is and whether it should live in data/ or not.