cbind_fill {str2str}

Bind DataFrames Along Columns - Filling in Missing Rows with NA

Description

cbind_fill binds together matrix-like objects by columns. The input objects will all internally be converted to data.frames by the generic function as.data.frame. When some objects do not contain rows that are present in other objects, NAs will be added to fill up the returned combined data.frame. If a matrix doesn't have rownames, the row number is used. Note that this means that a row with name "1" is merged with the first row of a matrix without name and so on. The returned matrix will always have row names. Colnames are ignored.

Usage

cbind_fill(...)

Arguments

...

any combination of data.frames, matrices, or atomic vectors input as separate arguments or a list.

Details

cbind_fill ensures each object has unique colnames and then calls Join(by = "0"). It is intended to be the column version of plyr::rbind.fill; it differs by allowing inputs to be matrices or vectors in addition to data.frames.

Value

data.frame created by combining all the objects input together. It will always have rownames. If colnames are not provided to the matrix-like objects, the returned colnames can be rather esoteric since default colnaming will be revised to ensure each colname is unique. If ... is a list of vectors, then the colnames will be the names of the list.

See Also

cbind_fill_matrix rbind.fill

Examples

# standard use
A <- data.frame("first" = 1:2, "second" = 3:4)
B <- data.frame("third" = 6:8, "fourth" = 9:11)
print(A)
print(B)
cbind_fill(A, B)

# help with unstack()
row_keep <- sample(1:nrow(InsectSprays), size = 36)
InsectSprays2 <- InsectSprays[row_keep, ]
unstacked <- unstack(InsectSprays2)
cbind_fill(unstacked)

# using rownames for binding
rownames(A) <- c("one", "two")
rownames(B) <- c("three","two","one")
print(A)
print(B)
cbind_fill(A, B)

# matrices as input
A <- matrix(1:4, 2)
B <- matrix(6:11, 3)
print(A)
print(B)
cbind_fill(A, B)

# atomic vector input
A <- data.frame("first" = 1:2, "second" = 3:4)
B <- data.frame("third" = 6:8, "fourth" = 9:11)
C <- c(12,13,14,15)
D <- c(16,17,18,19)
cbind_fill(A, B, C, D)

# same as plyr::rbind.fill, it doesn't handles well some inputs with custom rownames
# and others with default rownames
rownames(A) <- c("one", "two")
print(A)
print(B)
cbind_fill(A, B)