16 %in%

16.1 vector1 %in% vector2

################################################################.
################################################################.
##
## vector1 %in% vector2
##
## Returns a logical vector that is the same length as vector1.
## For every value in vector1 TRUE indicates that the value
## is in vector2 and FALSE indicates that it isn't.
## See examples below.
##
################################################################.
################################################################.

students <- c("joe", "sue", "al", "anne")
students

[1] "joe"  "sue"  "al"   "anne"

"sue" %in% students    # TRUE

[1] TRUE

"amanda" %in% students   # FALSE

[1] FALSE

c("sue","amanda") %in% students     # TRUE FALSE

[1]  TRUE FALSE

c("sue","amanda","felix","al","joaquin","xavier","zeke","anne")  %in% students

[1]  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE

# ANSWER :  TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE

16.2 — Practice —

#---------------------------------------------------------------------.
# QUESTION
#
# The variables mathClass and englishClass contain names of students
# who are in each respective class.
#
# The "built-in" intersect function returns the values that are
# common to two different vectors. ( see ?intersect )
#
#   (a) use the intersect function to show those students who are 
#       in both the math and the english class.
#
#   (b) do the same as (a) but do NOT use the intersect function,
#       rather use the %in% operator to accomplish the same result.
#
#   (c) write a function name myintersect that does the same
#       thing as the built-in intersection function - but do not
#       call the built-in intersection function in the 
#       definition of myintersection.
#---------------------------------------------------------------------.

# ANSWER

# Let's setup some data to work with
mathClass   = c("alice", "mike", "sue", "bob", "amanda")
englishClass = c("sue", "joe", "mike", "frank")

#   (a) use the intersect function to show those students who are 
#       in both the math and the english class.

# Let's look at the R documentation for the intersect function
?intersect

starting httpd help server ... done

# Use the function
intersect(mathClass, englishClass)

[1] "mike" "sue"

#   (b) do the same as (a) but do NOT use the intersect function,
#       rather use the %in% operator to accomplish the same result.
#
#       HINT - remember that %in% returns logical values (i.e. TRUE/FALSE values)
#              Use these TRUE/FALSE values to index into the appropriate vector.



# step 1 - figure out how to get a logical vector to show TRUE for each student
#          in the math class that is also in the english class

mathClass %in% englishClass

[1] FALSE  TRUE  TRUE FALSE FALSE

# step 2 - use the result from step1 to get the names of those students from the math class
#          who are also in the english class

mathClass [ mathClass %in% englishClass ]

[1] "mike" "sue"

# another way ....

englishClass %in% mathClass

[1]  TRUE FALSE  TRUE FALSE

englishClass[ englishClass %in% mathClass  ]

[1] "sue"  "mike"

#############################################################################.
# DON'T DO THE FOLLOWING! IT IS WRONG!
#############################################################################.
# The mathClass variable is used outside the brackets while
# the expression "englishClass %in% mathClass" gives one TRUE/FALSE for 
# each value in the englishClass variable. 
# You must make sure that the variable used outside the [brackets] matches up
# with the TRUE/FALSE values that are generated inside the [brackets].
#############################################################################.

mathClass[ englishClass %in% mathClass  ]  # THIS IS WRONG!!!

[1] "alice"  "sue"    "amanda"

# WHAT'S THE PROBLEM - FIGURE OUT WHAT HAPPENED
#
# original:           mathClass[ englishClass %in% mathClass  ]
#
# replace variables:  mathClass[c("sue","joe","mike","frank") %in% c("alice","mike","sue","bob","amanada")]
#
# %in%:               mathClass[c(TRUE, FALSE, TRUE, FALSE) ]
#
# recycle T/F values: mathClass[c(TRUE,FALSE,TRUE,FALSE,TRUE)]      
#
# final values:       "alice" "sue" "amanda"




#   (c) write a function name myintersect that does the same
#       thing as the built-in intersect function - but do not
#       call the built-in intersect function in the 
#       definition of myintersect.

myintersect = function(vec1, vec2) {
  
  vec1 [ vec1 %in% vec2 ]
  
}


myintersect( mathClass, englishClass)   # mike sue

[1] "mike" "sue"

myintersect( englishClass, mathClass)  # "sue" "mike"

[1] "sue"  "mike"

myintersect( 1:10, 7:1000) # 7 8 9 109

[1]  7  8  9 10

# Other ways ... 

myintersect = function(vec1, vec2) {
  return ( vec1 [ vec1 %in% vec2 ] )
}

myintersect = function(vec1, vec2) {
  tf = vec1 %in% vec2 
  vec1 [ tf ]
}

myintersect = function(vec1, vec2) {
  tf = vec1 %in% vec2 
  return ( vec1 [ tf ] )
}

#---------------------------------------------------------------------.
# QUESTION  ####
#
# The variables mathClass and englishClass contain names of students
# who are in each respective class.
#
# Write a command to show the students who are in the math class but
# are NOT in the english class
#
# (a) use the built-in setdiff function (read the documentation)
#
# (b) Do the same thing as (a) but WITHOUT using the setdiff function
#     HINT: use the ! operator and the %in% operator
#---------------------------------------------------------------------.

mathClass   = c("alice", "mike", "sue", "bob", "amanda")
englishClass = c("sue", "joe", "mike", "frank")

# ANSWER

# (a) use the built-in setdiff function (read the documentation)

setdiff(mathClass, englishClass)

[1] "alice"  "bob"    "amanda"

# (b) Do the same thing as (a) but WITHOUT using the setdiff function
#
#     HINT: use the ! operator and the %in% operator
#           Remember that ! will flip FALSE to TRUE and TRUE to FALSE.
#
#           > ! c(TRUE, FALSE, FALSE)
#           [1] FALSE TRUE TRUE

mathClass  %in% englishClass    # TRUE if someone is in Math and also in English     FALSE if they are just in math

[1] FALSE  TRUE  TRUE FALSE FALSE

! ( mathClass  %in% englishClass )  # FALSE if someone is in Math and also in English   TRUE if they are just in math

[1]  TRUE FALSE FALSE  TRUE  TRUE

mathClass [ ! ( mathClass  %in% englishClass )   ]

[1] "alice"  "bob"    "amanda"

16.3 ! SOME_VECTOR %in% ANOTHER_VECTOR

####################################################################################.
# Understand the placement of !  ####
#
# ! must immediately precede a logical vector (i.e. a vector of TRUE/FALSE values)
####################################################################################.

# Correct - ! operates on TRUE/FALSE values (AKA logical vectors)
! c(TRUE, FALSE)

[1] FALSE  TRUE

# Correct - parentheses contain TRUE/FALSE values
#           since %in% results in a logical vector
! ( mathClass  %in% englishClass )

[1]  TRUE FALSE FALSE  TRUE  TRUE

# Correct - same as above - %in% has higher precedence
#           than ! so you don't need (parentheses)
!   mathClass  %in% englishClass

[1]  TRUE FALSE FALSE  TRUE  TRUE

# Wrong - ! doesn't work with character values
! c("mike", "sue" )  # ERROR

Error in !c("mike", "sue"): invalid argument type

# Wrong - ! is in wrong place
mathClass ! %in% englishClass  # ERROR

Error: <text>:2:11: unexpected '!'
1: # Wrong - ! is in wrong place
2: mathClass !
             ^

16.4 — Practice —

#---------------------------------------------------------------------.
# QUESTION
#
# The variables mathClass and englishClass contain names of students
# who are in each respective class.
#
# (a) Write a command to show the students who are only in one of the classes.
#     Do not repeat the same name twice. HINT: use the setdiff function
#
# (b) do the same, but this time, do NOT use the setdiff function.
#---------------------------------------------------------------------.

mathClass   = c("alice", "mike", "sue", "bob", "amanda")
englishClass = c("sue", "joe", "mike", "frank")

# ANSWER - using the setdiff function and c function

onlyMath = setdiff(mathClass, englishClass)
onlyEnglish = setdiff(englishClass, mathClass)
onlyOneClass = c(onlyMath, onlyEnglish)

mathClass

[1] "alice"  "mike"   "sue"    "bob"    "amanda"

englishClass

[1] "sue"   "joe"   "mike"  "frank"

onlyMath

[1] "alice"  "bob"    "amanda"

onlyEnglish

[1] "joe"   "frank"

onlyOneClass

[1] "alice"  "bob"    "amanda" "joe"    "frank"

# all in one line 
c ( setdiff(mathClass, englishClass) , setdiff(englishClass, mathClass))

[1] "alice"  "bob"    "amanda" "joe"    "frank"

# ANSWER - do NOT use the setdiff function - You can split up the code into
#          different commands to make it easier to read.

onlyMath = mathClass [ ! mathClass %in% englishClass ]
onlyEnglish = englishClass [ ! englishClass %in% mathClass ]
c(onlyMath, onlyEnglish)

[1] "alice"  "bob"    "amanda" "joe"    "frank"