A New Week!

We aren’t computer scientists and that’s okay!

We make lots of mistakes. Mistakes are funny. You can laugh with us.

Let’s go, Simba, Pumbaa, and Timon!

1 string basics

We will focus on the stringr package for string manipulation.

All functions from stringr start with str_.

stringr cheatsheet

https://evoldyn.gitlab.io/evomics-2018/ref-sheets/R_strings.pdf

1.1 string length

Use str_length() to find the number of characters in a string.

library(tidyverse)

x <- c("a", "R is fun", NA)

str_length(x)

length(x)

str_count(x)

1.2 combine strings

Use str_c() to combine two or more strings

a <- 1

str_c("./output/aq-", a, ".csv")

str_c(10, 7, 2020, sep = "/")

# if input is a vector
x <- c("a", "b")

str_c(x, collapse = "") # specify collapse to combine input vectors into a single string

# two or more vectors
y <- c("1", "2", "3", "4") 

str_c(x, y) # the short vector is recycled to the length of the longest

1.3 subsetting strings

Use str_sub() to extract parts of a string based on index.

x <- fruit[c(1, 16, 10)]

str_sub(x, 1, 3)

str_sub(x, -3, -1)

1.4 convert case

str_to_lower() & str_to_upper()

str_to_lower(x)

str_to_upper(x)

1.5 sort strings

str_sort()

str_sort(x)

2 regular expressions

Regular expressions, or regexps, are a concise language for describing patterns in strings.

In R, we write regular expressions as strings.

We will use str_view() to learn regular expression

# install and load the htmlwidgets package 
install.packages("htmlwidgets") 
library(htmlwidgets)

# The simplest patterns match exact characters:
str_view(x, 'rry') # 'rry' is the pattern

# The next is special characters.
# For example, dot . matches any character (except a newline):
str_view(x, ".e.")

# But if . matches any character, how do you match the character .?

# Regexps use the backslash \ to escape special behaviour. So to match ., you need the \.

# However, whenever a \ appears in a regexp, you must write it as \\ in the string that represent the regexp. 

str_view(c("abc", "a.c", "bef"), "\.")

str_view(c("abc", "a.c", "bef"), "\\.")

# ^ to match the start of the string
str_view(x, "^a")

# $ to match the end of the string.
str_view(x, "y$")

#To remember which is which, try this: if you begin with power (^), you end up with money ($).

x <- c("apple pie", "apple", "apple cake")

str_view(x, "^apple$")

For more patterns with regular expressions, please check the stringr cheatshet.

3 match strings

3.1 str_extract() and str_extract_all()

To extract the actual text of a match and return as a list.

report <- "The first emission test was conducted on 10-09-2020 for three times. The second was tested on 10-11-2020. "

str_extract(report, "\\d{2}-\\d{2}-\\d{4}") # extract the first match

output <- str_extract_all(report, "\\d{2}-\\d{2}-\\d{4}") # extract all matches

3.2 str_detect()

To determine if a character vector matches a pattern. It returns a logical vector.

# basic usage
str_detect(string, pattern)

match <- str_detect(x, "erry")

You can also combine str_detect() and filter() to select rows in a dataframe.

us_rent_income %>% 
  mutate(NAME = str_to_lower(NAME)) %>%  # convert state name to lower case
  filter(str_detect(NAME, "ca")) # filter state names with "ca"

3.3 str_replace() and str_replace_all()

Replace matches with new strings

x <- c("appl e", "pear", "banana")

str_replace(x, " ", "")

x <- c("1 house", "2 cars", "3 people")

str_replace_all(x, c("1" = "one", "2" = "two", "3" = "three"))

3.4 str_split

Use str_split() to split a string up into pieces.

filename <- "D:/LearnR/input/ozone_samples_demo.csv"

output <- str_split(filename, '/')
output[[1]][4] # extract the file name

output1 <- unlist(output) # use unlist() to produce a vector from a list
output1[4]

Exercise

Extract date from a pdf file.

Obtain the pdf text using the following code.

pdf_file <- "https://cran.r-project.org/web/packages/pdftools/pdftools.pdf"

library(pdftools)
text <- pdf_text(pdf_file)

4 Recap

string basics
regular expressions
match strings