This function removes strings from a vector that are present in another vector.
Value
A vector containing the strings from the original vector that were not present in the removal vector.
Details
A set-difference helper for variable vectors, equivalent to
setdiff(Orig, Remove) but keeping duplicates in Orig. It exists for the
common case of taking an auto-detected variable set and dropping the few
columns that should not be analyzed - identifiers, dates, the outcome
itself - without retyping the ones that should.
Names in Remove that do not appear in Orig are ignored silently, so a
stale exclusion list will not error; it will simply stop having an effect.
See also
getNumVars(), getCatVars(), and getBinaryVars(), which
produce the vectors this usually trims.
Examples
data(SampleData)
data(SampleVariableTypes)
Labelled <- RevalueData(SampleData, SampleVariableTypes)$RevaluedData
# Every numeric column, including ones that are not really measurements
vars_Numeric <- getNumVars(Labelled)
length(vars_Numeric)
#> [1] 128
# Drop the ones that should not be screened as biomarkers
vars_Biomarkers <- removeString(vars_Numeric, c("age", "tau", "p_tau"))
length(vars_Biomarkers)
#> [1] 125
utils::head(vars_Biomarkers, 4)
#> [1] "ACE_CD143_Angiotensin_Converti" "ACTH_Adrenocorticotropic_Hormon"
#> [3] "AXL" "Adiponectin"
# Names that are not present are ignored, so an over-broad exclusion list
# is harmless
removeString(c("age", "sex", "AXL"), c("sex", "not_a_column"))
#> [1] "age" "AXL"
# Unlike setdiff(), duplicates in the original are kept
removeString(c("a", "a", "b", "c"), "b")
#> [1] "a" "a" "c"
setdiff(c("a", "a", "b", "c"), "b")
#> [1] "a" "c"