rBind.fill: Combine matrixes by row, fill in missing columns

Description

rbinds a list of Matrix or matrix like objects, filling in missing columns.

Usage

rBind.fill(x, ..., fill = NULL, out.class = class(rbind(x, x))[1])

Arguments

x, ...

Objects to combine. If the first argument is a list and .. is unpopulated, the objects in that list will be combined.

fill

value with which to fill unmatched columns

out.class

the class of the output object. Defaults to the class of x. Note that some output classes are not possible due to R coercion capabilities, such as converting a character matrix to a Matrix.

Value

a single object of the same class as the first input, or of class matrix if the first object is one dimensional

Details

Similar to rbind.fill.matrix, but works for Matrix as well as all other R objects. It is completely agnostic to class, and will produce an object of the class of the first input (or of class matrix if the first object is one dimensional).

The implementation is recursive, so it can handle an arbitrary number of inputs, albeit inefficiently for large numbers of inputs.

This method is still experimental, but should work in most cases. If the data sets consist solely of data frames, rbind.fill is preferred.

Examples

Run this code

# NOT RUN {
df1 = data.frame(x = c(1,2,3), y = c(4,5,6))
rownames(df1) = c("a", "b", "c")
df2 = data.frame(x = c(7,8), z = c(9,10))
rownames(df2) = c("a", "d")
rBind.fill(df1,df2,fill=NA)
rBind.fill(as(df1,'Matrix'),df2,fill=0)
rBind.fill(as.matrix(df1),as(df2,'Matrix'),c(1,2),fill=0)
rBind.fill(c(1,2,3),list(4,5,6,7))
rBind.fill(df1,c(1,2,3,4))

m<-rsparsematrix(1000000,100,.001)
m2<-m
colnames(m)<-1:100
colnames(m2)<-3:102
system.time(b<-rBind.fill(m,m2))

# }