replyr_mapRestrictCols: Map names of columns to known values and drop other columns.
In WinVector/replyr: Patches to Use 'dplyr' on Remote Data Sources

Description Usage Arguments Details Value See Also Examples

Restrict a data item's column names and re-name them in bulk.

1	replyr_mapRestrictCols(x, nmap, ..., restrict = FALSE, reverse = FALSE)

`x`	data item to work on
`nmap`	named list mapping with keys specifying new column names, and values as original column names.
`...`	force later arguments to bind by name
`restrict`	logical if TRUE restrict to columns mentioned in nmap.
`reverse`	logical if TRUE apply the inverse of nmap instead of nmap.

Something like replyr::replyr_mapRestrictCols is only useful to get control of a function that is not parameterized (in the sense it has hard-coded column names inside its implementation that don't the match column names in our data).

data item with columns renamed (and possibly restricted).

let, replyr_reverseMap, replyr_apply_f_mapped

# an external function with hard-coded column names
DecreaseRankColumnByOne <- function(d) {
  d$RankColumn <- d$RankColumn - 1
  d
}

# our example data, with different column names
d <- data.frame(Sepal_Length=c(5.8,5.7),
                Sepal_Width=c(4.0,4.4),
                Species='setosa',rank=c(1,2))
print(d)


# map our data to expected column names so we can use function
nmap <- c(GroupColumn='Species',
          ValueColumn='Sepal_Length',
          RankColumn='rank')
print(nmap)
dm <- replyr_mapRestrictCols(d,nmap)
print(dm)

# can now apply code that expects hard-coded names.
dm <- DecreaseRankColumnByOne(dm)

# map back to our original column names (for the columns we retained)
# Note: can only map back columns that were retained in first mapping.
replyr_mapRestrictCols(dm, nmap, reverse=TRUE)