Arrange univariateTable output by values not by levels - r

Question

I am trying to solve the following inconvenience when trying to export a table consisting of factor levels. Here is the code to generate the sample data, and a table from it.
data <- c(sample('A',30,replace=TRUE), sample('B',120,replace=TRUE),
sample('C',180,replace=TRUE), sample('D',70,replace=TRUE))
library(Publish)
univariateTable(~data)
The default output of the univariateTable is by levels (From A through D):
Variable Levels Value
1 data A 30 (7.5)
2 B 120 (30.0)
3 C 180 (45.0)
4 D 70 (17.5)
How can I change this so that the output is based on the value instead? I mean, the first row being the largest number (and percentage) and the last low being the lowest, like this:
Variable Levels Value
1 data C 180 (45.0)
2 B 120 (30.0)
3 D 70 (17.5)
4 A 30 (7.5)

Answer 1

Assuming that the "Publish" package is the one installed from github, we extract the numbers before the ( using sub, order it and use it to order the "xlevels" and "summary.totals".
#library(devtools)
#install_github("TagTeam/Publish")
library(Publish)
Out <- univariateTable(~data)
i1 <- order(as.numeric(sub('\\s+.*', '',
Out$summary.totals$data)), decreasing=TRUE)
Out$xlevels$data <- Out$xlevels$data[i1]
Out$summary.totals$data <- Out$summary.totals$data[i1]
Out
# Variable Level Total
#1 data C 180 (45.0)
#2 B 120 (30.0)
#3 D 70 (17.5)
#4 A 30 (7.5)
data
set.seed(24)
data <- c(sample('A',30,replace=TRUE), sample('B',120,replace=TRUE),
sample('C',180,replace=TRUE), sample('D',70,replace=TRUE))