Mirror Matrix Dimensional Member Functions for DataFrame

[coatless]

The proposal here is to improve API consistency between the 2D classes Matrix and DataFrame by:

1. deprecating the .nrows() function in DataFrame in favor of adding .nrow() and .rows()
2. introduce the .ncol() and .cols() functions to DataFrame that map to Vector's size() and length() functions.

However, another interesting proposal would be to remove the size attribute access information provided by .cols() and .rows() for the Matrix class and introduce .Rows(first, last) and .Cols(first, last) that would provide submatrix views.

APIs

For the Matrix class, the following member functions are defined for dimensional information:

Member	Description
.nrow(),.rows()	number of rows in a Matrix
.ncol(),.cols()	number of columns in a Matrix
.size(),.length()	number of items in a Matrix, Vector

Meanwhile, over in DataFrame land, the member functions are defined a bit differently:

Member	Description
.nrows()	number of rows in a DataFrame
.size(),.length()	number of columns in a DataFrame

The break in consistency regarding the nrow() and ncol() within DataFrame is problematic as it breaks with those accessor functions available in base R.

data.frame and length() note

Note, there is no need to change the length() function call since the R equivalent returns the number of columns instead of the amount of elements, e.g.

df = data.frame(a = 1:3, b = 3:1)
length(df)
# 2