Migration note: argument renaming in 19.15.0

library(SciDataReportR)

SciDataReportR 19.15.0 standardized argument names across the whole package to snake_case. Function names did not change (still PascalCase), argument positions did not change, and every old argument name still works - it just emits one deprecation warning per call pointing to the new name.

What was renamed

The same concept used to go by several names in different functions. Each cluster now has one name everywhere:

Concept	Old names (varied by function)	New name
Primary data frame	Data, DataFrame, df, Dataframe, DatatoRevalue	data
Variable selection	Variables, variables	variables
Covariates	Covariates, Covars, covars, covariates	covariates
Merge keys	Keys, MergeBy, key	keys (safe_merge() keeps by to mirror dplyr::left_join())
Participant ID column	ID, id, id_col	id_var
Grouping variable	GroupVariable, GroupVar, Group, CompVariable	group_var
Codebook	Codebook, CB, VariableCodebook, VarTypes	codebook
Digits	numdecimals	digits
Digits (multi-knob tables)	ValueDigits, pDigits, EffectSizeDigits, TooltipDigits	value_digits, p_digits, effect_size_digits, tooltip_digits
Interactive output	Interactive, MakeInteractive, make_interactive	interactive (each function keeps its original default)
Predictors (directional analyses)	xVars, PredictorVars, predictorVars, predictorVar	predictor_vars / predictor_var
Outcomes (directional analyses)	yVars, yVar, OutcomeVars, outcomeVars, outcomeVar	outcome_vars / outcome_var
Symmetric variable set	xVars/yVars in PlotDirectionalHeatmaps()	variables

Also new in 19.15.0: CreatePCAObject()’s duplicate imputeMethod argument is deprecated in favor of ImputeMethod, and the heatmap/matrix family gained an fdr_scope argument (see the FDR correction scope vignette).

What you will see

Old code keeps working. The first thing you’ll notice is a one-line warning per deprecated argument, telling you the replacement:

old_style <- MakeDataDictionary(DataFrame = mtcars, numdecimals = 2)
#> Warning: The `DataFrame` argument of `MakeDataDictionary()` is deprecated as of
#> SciDataReportR 19.15.0.
#> ℹ Please use the `data` argument instead.
#> Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
#> generated.
#> Warning: The `numdecimals` argument of `MakeDataDictionary()` is deprecated as of
#> SciDataReportR 19.15.0.
#> ℹ Please use the `digits` argument instead.
#> Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
#> generated.

The new-style call is silent and returns the identical result:

new_style <- MakeDataDictionary(mtcars, digits = 2)
identical(old_style, new_style)
#> [1] TRUE

Positional calls never changed, because every argument kept its position (the deprecated names were appended at the end of each signature):

identical(MakeDataDictionary(mtcars, 2), new_style)
#> [1] TRUE

# a directional example: xVars/yVars became predictor_vars/outcome_vars
res <- PlotCorrelationsHeatmap(
  mtcars,
  predictor_vars = c("hp", "wt"),
  outcome_vars   = "mpg"
)
round(res$Unadjusted$p, 4)
#>    mpg
#> hp   0
#> wt   0

Migrating

There is no urgency - deprecated arguments will keep working throughout the 19.x series. To migrate a script, rename arguments per the table above; nothing else changes. If you want deprecation warnings to be errors while you migrate (to catch every old usage), set:

options(lifecycle_verbosity = "error")