-Dealing with String

Handling, cleaning and processing character strings is becoming a prerequisite in daily data analysis.

#Character string basics

Creating Strings:

x <- "Berlin"    
y <- "Paris"     

#The paste() function is expert for creating text

# paste together string a & b
paste(a, b)                      
## [1] "learning to create character strings"

# paste character and number strings 
# (converts numbers to character class)
paste("The life of", pi)           
## [1] "The life of 3.14159265358979"

# paste multiple strings
paste("I", "love", "R")            
## [1] "I love R"

# paste multiple strings with a separating character
paste("I", "love", "R", sep = "-")  
## [1] "I-love-R"

# use paste0() to paste without spaces btwn characters
paste0("I", "love", "R")            
## [1] "IloveR"

# paste objects with different lengths
paste("R", 1:5, sep = " v1.")       
## [1] "R v1.1" "R v1.2" "R v1.3" "R v1.4" "R v1.5"

Converting to Strings:

a <- "The life of"    
b <- pi

is.character(a)
## [1] TRUE

is.character(b)
## [1] FALSE

c <- as.character(b)
is.character(c)
## [1] TRUE

toString(c("Aug", 24, 1980))
## [1] "Aug, 24, 1980"

Printing Strings:

• print(): generic printing

• noquote(): print with no quotes

• cat(): concatenate and print with no quotes

• sprintf(): a wrapper for the C function sprintf, that returns a character vector containing a formatted combination of text and variable values

x <- "learning to print strings"    

# basic printing
print(x)                
## [1] "learning to print strings"

# print without quotes
print(x, quote = FALSE)  
## [1] learning to print strings

noquote(x)
## [1] learning to print strings

#Alternative:
# basic printing (similar to noquote)
cat(x)                   
## learning to print strings

# combining character strings
cat(x, "in R")           
## learning to print strings in R

# basic printing of alphabet
cat(letters)             
## a b c d e f g h i j k l m n o p q r s t u v w x y z

# specify a seperator between the combined characters
cat(letters, sep = "-")  
## a-b-c-d-e-f-g-h-i-j-k-l-m-n-o-p-q-r-s-t-u-v-w-x-y-z

# collapse the space between the combine characters
cat(letters, sep = "")   
## abcdefghijklmnopqrstuvwxyz

x <- "Today I am learning how to print strings."
y <- "Tomorrow I plan to learn about textual analysis."
z <- "The day after I will take a break and drink a beer."

cat(x, y, z, fill = 0)
## Today I am learning how to print strings. ->
# Tomorrow I plan to learn about textual analysis. ->
# The day after I will take a break and drink a beer.

cat(x, y, z, fill = 5)
## Today I am learning how to print strings. 
## Tomorrow I plan to learn about textual analysis. 
## The day after I will take a break and drink a beer.

To substitute in a string or string variable, use %s:

x <- "print strings"

# substitute a single string/variable
sprintf("Learning to %s in R", x)    
## [1] "Learning to print strings in R"

# substitute multiple strings/variables
y <- "in R"
sprintf("Learning to %s %s", x, y)   
## [1] "Learning to print strings in R"

For integers, use %d or a variant:

version <- 3

# substitute integer
sprintf("This is R version:%d", version)
## [1] "This is R version:3"

# print with leading spaces
sprintf("This is R version:%4d", version)   
## [1] "This is R version:   3"

# can also lead with zeros
sprintf("This is R version:%04d", version)   
## [1] "This is R version:0003"

For floating-point numbers, use %f for standard notation, and %e or %E for exponential notation:

sprintf("%f", pi)         # '%f' indicates 'fixed point' decimal 
#notation
## [1] "3.141593"

sprintf("%.3f", pi)       # decimal notation with 3 decimal digits
## [1] "3.142"

sprintf("%1.0f", pi)      # 1 integer and 0 decimal digits
## [1] "3"

sprintf("%5.1f", pi)      # decimal notation with 5 total 
#decimal digits  
## [1] "  3.1"            # only 1 to the right of the decimal point

sprintf("%05.1f", pi)     # same as above but fill empty digits 
#with zeros
## [1] "003.1"

sprintf("%+f", pi)        # print with sign (positive)
## [1] "+3.141593"

sprintf("% f", pi)        # prefix a space
## [1] " 3.141593"

sprintf("%e", pi)         # exponential decimal notation 'e'
## [1] "3.141593e+00"

sprintf("%E", pi)         # exponential decimal notation 'E'
## [1] "3.141593E+00"

Counting string elements and characters:

length("How many elements are in this string?")
## [1] 1

length(c("How", "many", "elements", "are", "in", "this", "string?"))
## [1] 7

nchar("How many characters are in this string?")
## [1] 39

nchar(c("How", "many", "characters", "are", "in", "this", "string?"))
## [1]  3  4 10  3  2  4  7

#String manipulation with base R

Case conversion:

x <- "Learning To MANIPULATE strinGS in R"

tolower(x)
## [1] "learning to manipulate strings in r"

toupper(x)
## [1] "LEARNING TO MANIPULATE STRINGS IN R"

Simple Character Replacement:

# replace 'A' with 'a'
x <- "This is A string."
chartr(old = "A", new = "a", x)
## [1] "This is a string."

# multiple character replacements
# replace any 'd' with 't' and any 'z' with 'a'
y <- "Tomorrow I plzn do lezrn zbout dexduzl znzlysis."
chartr(old = "dz", new = "ta", y)
## [1] "Tomorrow I plan to learn about textual analysis."

String Abbreviations:

streets <- c("Main", "Elm", "Riverbend", "Mario", "Frederick")

# default abbreviations
abbreviate(streets)
##      Main       Elm Riverbend     Mario Frederick 
##    "Main"     "Elm"    "Rvrb"    "Mari"    "Frdr"

# set minimum length of abbreviation
abbreviate(streets, minlength = 2)
##      Main       Elm Riverbend     Mario Frederick 
##      "Mn"      "El"      "Rv"      "Mr"      "Fr"

Extract/Replace Substrings:

alphabet <- paste(LETTERS, collapse = "")

# extract 18th character in string
substr(alphabet, start = 18, stop = 18)
## [1] "R"

# extract 18-24th characters in string
substr(alphabet, start = 18, stop = 24)
## [1] "RSTUVWX"

# replace 1st-17th characters with `R`
substr(alphabet, start = 19, stop = 24) <- "RRRRRR"
alphabet
## [1] "ABCDEFGHIJKLMNOPQRRRRRRRYZ"

alphabet <- paste(LETTERS, collapse = "")

# extract 18th through last character
substring(alphabet, first = 18)
## [1] "RSTUVWXYZ"

# recursive extraction; specify start position only
substring(alphabet, first = 18:24)
## [1] "RSTUVWXYZ" "STUVWXYZ"  
# "TUVWXYZ"   "UVWXYZ"    "VWXYZ"     "WXYZ"     
## [7] "XYZ"

# recursive extraction; specify start and stop positions
substring(alphabet, first = 1:5, last = 3:7)
## [1] "ABC" "BCD" "CDE" "DEF" "EFG"

z <- "The day after I will take a break and drink a beer."
strsplit(z, split = " ")
## [[1]]
##  [1] "The"   "day"   "after" "I"     "will"  "take"  "a"     "break"
##  [9] "and"   "drink" "a"     "beer."

a <- "Alabama-Alaska-Arizona-Arkansas-California"
strsplit(a, split = "-")
## [[1]]
## [1] "Alabama"    "Alaska"     "Arizona"    "Arkansas"   "California"

unlist(strsplit(a, split = "-"))
## [1] "Alabama"    "Alaska"     "Arizona"    "Arkansas"   "California"

#String manipulation with stringr

The stringr package is the Winner problem solving for string manipulation. Developer by Hadley Wickham

# install stringr package
install.packages("stringr")

# load package
library(stringr)

Basic Operations:

Three string functions that are closely related to their base R equivalents:

• Concatenate with str_c()

• Number of characters with str_length()

• Substring with str_sub()

# same as paste0()
str_c("Learning", "to", "use", "the", "stringr", "package")
## [1] "Learningtousethestringrpackage"

# same as paste()
str_c("Learning", "to", "use", "the", "stringr", "package", sep = " ")
## [1] "Learning to use the stringr package"

# allows recycling 
str_c(letters, " is for", "...")
##  [1] "a is for..." "b is for..." "c is for..."......

# some text with NA
text = c("Learning", "to", NA, "use", "the", NA, "stringr", "package")

# compare `str_length()` with `nchar()`
nchar(text)
## [1] 8 2 2 3 3 2 7 7

str_length(text)
## [1]  8  2 NA  3  3 NA  7  7

x <- "Learning to use the stringr package"

# alternative indexing
str_sub(x, start = 1, end = 15)
## [1] "Learning to use"

str_sub(x, end = 15)
## [1] "Learning to use"

str_sub(x, start = 17)
## [1] "the stringr package"

str_sub(x, start = c(1, 17), end = c(15, 35))
## [1] "Learning to use"     "the stringr package"

# using negative indices for start/end points from end of string
str_sub(x, start = -1)
## [1] "e"

str_sub(x, start = -19)
## [1] "the stringr package"

str_sub(x, end = -21)
## [1] "Learning to use"

# Replacement
str_sub(x, end = 15) <- "I know how to use"
x
## [1] "I know how to use the stringr package"

Duplicate Characters within a String:

str_dup("beer", times = 3)
## [1] "beerbeerbeer"

str_dup("beer", times = 1:3)
## [1] "beer"         "beerbeer"     "beerbeerbeer"


# use with a vector of strings
states_i_luv <- state.name[c(6, 23, 34, 35)]
str_dup(states_i_luv, times = 2)
## [1] "ColoradoColorado"         "MinnesotaMinnesota"      
## [3] "North DakotaNorth Dakota" "OhioOhio"

Remove Leading and Trailing Whitespace:

# remove whitespaces on the left side
str_trim(text, side = "left")
## [1] "Text "       "with"        "whitespace " "on"      "both "      
## [6] "sides "

# remove whitespaces on the right side
str_trim(text, side = "right")
## [1] "Text"        "  with"      " whitespace" " on"     "both"       
## [6] " sides"

# remove whitespaces on both sides
str_trim(text, side = "both")
## [1] "Text"       "with"       "whitespace" "on"         "both"      
## [6] "sides"

Pad a String with Whitespace:

str_pad("beer", width = 10, side = "left")
## [1] "      beer"

str_pad("beer", width = 10, side = "both")
## [1] "   beer   "

str_pad("beer", width = 10, side = "right", pad = "!")
## [1] "beer!!!!!!"

#Set operatons for character strings

Set Union:

set_1 <- c("lagunitas", "bells", "dogfish", "summit", "odell")
set_2 <- c("sierra", "bells", "harpoon", "lagunitas", "founders")

union(set_1, set_2)
## [1] "lagunitas" "bells"     "dogfish"   "summit"    
#"odell"     "sierra"   
## [7] "harpoon"   "founders"

Set Intersection:

intersect(set_1, set_2)
## [1] "lagunitas" "bells"

Identifying Different Elements:

# returns elements in set_1 not in set_2
setdiff(set_1, set_2)
## [1] "dogfish" "summit"  "odell"

# returns elements in set_2 not in set_1
setdiff(set_2, set_1)
## [1] "sierra"   "harpoon"  "founders"

Testing for Element Equality:

set_3 <- c("woody", "buzz", "rex")
set_4 <- c("woody", "andy", "buzz")
set_5 <- c("andy", "buzz", "woody")

setequal(set_3, set_4)
## [1] FALSE

setequal(set_4, set_5)
## [1] TRUE

Testing for Exact Equality:

set_6 <- c("woody", "andy", "buzz")
set_7 <- c("andy", "buzz", "woody")
set_8 <- c("woody", "andy", "buzz")

identical(set_6, set_7)
## [1] FALSE

identical(set_6, set_8)
## [1] TRUE

Identifying if Elements are Contained in a String:

good <- "andy"
bad <- "sid"

is.element(good, set_8)
## [1] TRUE

good %in% set_8
## [1] TRUE

bad %in% set_8
## [1] FALSE

Sorting a String:

sort(set_8)
## [1] "andy"  "buzz"  "woody"

sort(set_8, decreasing = TRUE)
## [1] "woody" "buzz"  "andy"