[feat] As client, I'd like part 1 of the flat_main to be a documented and tested function

[statnmap]

As client, I'd like part 1 of the flat_main to be a documented and tested function, so that I can guarantee its outputs, whatever changes in the packages

Client - Validation

• [x] There is a function with all parameters needed, which returns outputs of part 1 as one object (a list)

...

=> https://github.com/balglave/FishMap/blob/main/R/fm_load_data.R

• [x] This function is tested, at least to be sure there no error (it tests that rds file outputs exists)

=> https://github.com/balglave/FishMap/blob/main/tests/testthat/test-fm_load_data.R

• [x] This function is used in the main vignette

=> https://balglave.github.io/FishMap/articles/running-fishmap.html

Dev - Tech

• [x] Transform code of the anonymous chunk as a function: Use the RStudio tool to do it:
  • Select code as is (without function()) => Go to Menu "Code" > Extract Function. Give the function a name, all variables should be extracted. You can try with a simple example like: print(toto)
• [x] Get all necessary outputs as a list()
  • Discover outputs by running the next chunk, so that we only get necessary outputs.
• The lists of outputs

  outputs <- ma_function_prep(  return(list(nom = contenu, nom2 = contenu2)))
  saveRDS(outputs, "outputs_part1.rds") # in FISHMAP_OUTPUT
  toto <- outputs[["toto"]]

[dagousket]

The refactoring of part1 raised the following warning in the devtools::check()

W  checking whether package ‘FishMap’ can be installed (27.6s)
   Found the following significant warnings:
     Warning: replacing previous import ‘dplyr::union’ by ‘raster::union’ when loading ‘FishMap’
     Warning: replacing previous import ‘dplyr::select’ by ‘raster::select’ when loading ‘FishMap’
     Warning: replacing previous import ‘dplyr::intersect’ by ‘raster::intersect’ when loading ‘FishMap’
     Warning: replacing previous import ‘raster::extract’ by ‘tidyr::extract’ when loading ‘FishMap’

-> this is now fixed

[JCasemajor]

I would suggest putting the setup of the spatial domain (grid_xmin, grid_xmax, grid_ymin, grid_ymax) as parameter for the fm_load_data function as the user will probably change it according to its data (i.e. if we want to study Celtic sea instead of the Bay of Biscay).

@balglave what do you think ?

[balglave]

D'accord avec Juliette. Ces arguments sont necessaires dans 'domain_mesh_spde.R'.

[statnmap]

Merci. Nous avons ouvert un ticket dédié pour ces paramètres particuliers: https://github.com/balglave/FishMap/issues/34
Vous aurez un autre ticket ensuite pour lister tous les autres paramètres que vous jugez nécessaires.

Je ferme le présent ticket si c'est ok pour vous.

[JCasemajor]

Je valide

[balglave]

Je valide