Editing text

Rules for editing column names:

• All lower case when possible

• Name should be descriptive, not shortcuts

• No duplicates

• No underscores, dots or white spaces

Rules for values:

• Should be factors

• Should be descrpitive, like TRUE/FALSE instead of 1/0.

Get some data to edit

We will use data from the following page: https://data.baltimorecity.gov/Transportation/Baltimore-Fixed-Speed-Cameras/dz54-2aru

Run this code in order to download sample data for this tutorial.

if (!file.exists("./data")) { dir.create("./data") }
url <- "https://data.baltimorecity.gov/api/views/dz54-2aru/rows.csv?accessType=DOWNLOAD"

download.file(url, destfile="./data/cameras.csv", method="curl")

data <- read.csv("./data/cameras.csv")

names(data)

Lower or upper case

tolower(names(data))
toupper(names(data))

Split values

For example, split word when there is a dot.

strsplit(names(data), "\\.")

Get only first element

In this example, we will split a string and get only first element.

firstElement <- function (x) { unlist(strsplit(x, ".", fixed=TRUE))[1] }

sapply(as.list(names(data)), firstElement)

Replace characters

The following code will replace all _ by empty space.

sub("_", " ", names(reviews))

sub function will always replace only first found characeter. In order to replace all, use gsub.

> test <- "test_value_1"

> sub("_", " ", test)
[1] "test value_1"

> gsub("_", " ", test)
[1] "test value 1"

Searching

Search for a string in a column.

> grep("Alameda", data$intersection)
[1]  4  5 36

How many times a string appears in column.

> table(grepl("Alameda", data$intersection))

FALSE  TRUE
   77     3

More about grep function is here.

More string functions

We can use stringr package to easier work with strings.

install.packages("stringr")
library(stringr)

Get number of characters.

> nchar("Hello there")
[1] 11

Get sub string.

substr("Hello there", 1,5)

Connect strings into one.

> paste("Hello", "there")
[1] "Hello there"

> paste0("Hello", "there")
[1] "Hellothere"

Trim string.

str_trim("   Hello   ")