df.sorter: Sort a 'data.frame' by Rows or Columns

In mrdwab/mrdwabmisc: Miscellaneous R functions, mostly for data processing

Description

The df.sorter function allows you to sort a data.frame by columns or rows or both. You can also quickly subset data columns by using the var.order argument.

Usage

df.sorter(data, var.order = names(data), col.sort = NULL, at.start = TRUE)

Arguments

data	The source data.frame.
var.order	The new order in which you want the variables to appear. See Details
col.sort	The columns within which there is data that need to be sorted. See Details.
at.start	Should the pattern matching be from the start of the variable name? Defaults to TRUE.

Details

var.order

• Defaults to names(data), which keeps the variables in the original order.

• Variables can be referred to either by a vector of their index numbers or by a vector of the variable name; partial name matching also works, but requires that the partial match identifies similar columns uniquely (see Examples). Basic subsetting can also be done using var.order simply by omitting the variables you want to drop.

col.sort

• Defaults to NULL, which means no sorting takes place. Variables can be referred to either by a vector of their index numbers or by a vector of the variable names; full names must be provided.

Note

If you are sorting both by variables and within the columns and using numeric indexes as opposed to variable names, the col.sort order should be based on the location of the columns in the new data.frame, not the original data.frame.

Author(s)

Ananda Mahto

Examples

# Make up some data
set.seed(1)
dat = data.frame(id = rep(1:5, each=3), times = rep(1:3, 5),
            measure1 = rnorm(15), score1 = sample(300, 15),
            code1 = replicate(15, paste(sample(LETTERS[1:5], 3),
                                        sep="", collapse="")),
            measure2 = rnorm(15), score2 = sample(150:300, 15),
            code2 = replicate(15, paste(sample(LETTERS[1:5], 3),
                                        sep="", collapse="")))
# Preview your data
dat

# Change the variable order, grouping related columns
# Note that you do not need to specify full variable names,
#    just enough that the variables can be uniquely identified
head(df.sorter(dat, var.order = c("id", "ti", "cod", "mea", "sco")))

# As above, but sorted by 'times' and then 'id'
head(df.sorter(dat,
          var.order = c("id", "tim", "cod", "mea", "sco"),
          col.sort = c(2, 1)))

# Drop 'measure1' and 'measure2', sort by 'times', and 'score1'
head(df.sorter(dat,
          var.order = c("id", "tim", "sco", "cod"),
          col.sort = c(2, 3)))

# Just sort by columns, first by 'times' then by 'id'
head(df.sorter(dat, col.sort = c("times", "id")))

# Pattern matching anywhere in the variable name
head(df.sorter(dat, var.order= "co", at.start=FALSE))

mrdwab/mrdwabmisc documentation built on May 23, 2019, 7:15 a.m.