vincyanebadouard / logginglab Goto Github PK

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

To increase user-friendliness, I'm implementing new developments on personal branches of my fork (uf_basic, uf_intermediate, and uf-advance). The developments are:

Basic

• Remove checks for unused columns PlotArea
• Remove checks for unused columns VernName, Circ, Xfield, Yfield, Lat, Lon
• Check Plot column only if present
• Check CensusYear column only if present
• Ease basic types inputs with autoconversion tests (with as.integer, as.logical and as.character)
• Ease table types inputs with autoconversion (DT & tibble & data.frame)
• Ease spatial types inputs with autoconversion (raster & terra, sp & sf)
• Add a stop if check/prepare inputs remove all lines
• Add a message if DBH or diameter column exists but not Circ
• Check DBH interval (1cm to 10m)

Intermediate

• Replace creekdistances argument for a named list with two arguments for two raster inputs
• Add an helper function to creat the mask

Advance

• Replace LoggingSimulationOutputs and LoggingSimulationOutputs_iter by new R objects with associated class to simplify and consolidate outputs and ease package use with commonly known methods (loggingsummary & loggingsummary1 to summary, show, print, autoplot)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

example : "Paracou6_2016" you click on it and you arrive on the documentation of this data

inventorycheckformat

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

addtreedim

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

Vincyane question

• Testing data
• Default data
  • Species criteria (minimum diameter, economic interest...)
  • Global parameters (track size ...)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

If I have scenarios to simulate, and the parameters of these scenarios are the arguments of my function,
ex : scenario 1 : A = 1, B = 4, C = 10
manual: we choose the values of A, B and C
fun(scenario, A, B, C)

what is the best strategy?

• create a table from which the function will take the values of the arguments (overwriting those that the user could have put)

• lines of code at the beginning of each function to assign values to the arguments according to the chosen scenario

• put in the code the scenarios, and consider the arguments only in the "manual" mode
  @sylvainschmitt

Vincyane question

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

I think we need to discuss the generecity of Maria if we want to publish it. I can't do it in code review because it is global to the package. So for the moment I'll stop making remarks about it. And I think it is not an issue for the version 1 that you are planning for the ONF / contract end. But we should discuss it with Geraldine when she come back from holidays.

@thomasgaquiere , what is the purpose off the following lines:

preMainTrails <- DTMExtended > -Inf
preMainTrails<- rasterToPolygons(preMainTrails, dissolve=T)

Using a dummy raster of 20m height, this provoke a bug in my last install because of multiple polygons:

dem <- raster::raster(mask, nrow = 200, ncol = 200)
values(dem) <- rep(20, length(values(dem)))
maintrailextract(dem)
# Erreur dans st_cast.sfc(st_geometry(x), to, group_or_split = do_split) : 
# use smaller steps for st_cast; first cast to MULTILINESTRING or POLYGON?

This is because the multiple polygons can't be casted into one linestring here:

MainTrails <- preMainTrails %>% st_as_sf() %>% st_cast(to = 'LINESTRING', warn= FALSE)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

Vincyane question

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

seems to be a problem with sample (all trees are either success or failure)

parallel is a basic R pacakge and should be at the same version as R so if you ask for R>=3.6 parallel should also be 3.6:

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

The test test-getmask.R fails.

I changed it in the em development branch but this must be double checked: either it is OK so correct the documentation of the function or correct the error.

What command can I give to a user who would like to load my package now and use it, if my repository is private? @sylvainschmitt

I created a broken dataset to test my "datacheckformat" function (check & control my inventory dataset.
When I put my script creating this "broken dataset", in data-raw folder (with use_data_raw()),
usethis says me to add this data to the package with usethis::use_data().
For my part, I put this data in inst/extdata because I don't want to give acces to this data at users
What I had to do ? put this data in data folder or in inst/extdata? @sylvainschmitt

We need to define lowlands, to define the UPs.

Several possibilities :

• altitude relative à la crique
• TWI (TopoWetnessIndex)
  - Rsaga : Saga has to be downloaded (external dependence). We cannot force the user to download it, nor can we download it for him. Need to be optional. (possibly not all the TWI formulas)
  - reuse Saga's codes (all the formulas)
  - package "dynatopmodel" : only 1 TWI formule

Computation time (ONF, Guyafor) and necessary precision (SRTM/LIDAR) to be taken into account

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

ONF's version

• code
• test
• document (V0)
• Sylvain's check (code)
• Geraldine check (doc)

@sylvainschmitt