On this page:
For Loops
Pattern Matching

Intermezzo: Scope and Abstraction

While the preceding part gets away with explaining local and lambda in an informal manner, the introduction of such abstraction mechanisms really requires additional terminology to facilitate such discussions. In particular, these discussions need words to delineate regions within programs and to refer to specific uses of variables.

This intermezzo starts with a section that defines the new terminology, most importantly, scope, binding variables, and bound variables. It immediately uses this new capability to introduce two abstraction mechanisms often found in programming languages:While the 2htdp/abstraction library is useful for the remaining parts of the book, the latter do not rely on any of its features. Instructors may wish to use the library anyway, though they should also explain then how the principles of design apply to languages without such features. for loops and pattern matching. The former is an alternative to functions such as map, build-list, andmap, etc; the latter abstracts over the conditional in the functions of the first three parts of the book. Both require not only the definition of functions but the creation of entirely new language constructs, meaning they are not something programmers can usually design and add to their vocabulary.


Consider the following two definitions:
(define (f x) (+ (* x x) 25))
(define (g x) (+ (f (+ x 1)) (f (- x 1))))
Clearly, the occurrences of x in f are completely unrelated to the occurrences of x in the definition of g. We could systematically replace the shaded occurrences with y and the function would still compute the exact same result. In short, the shaded occurrences of x have meaning only inside the definition of f and nowhere else.

At the same time, the first occurrence of x in f is different from the others. When we evaluate (f n), the occurrence of f completely disappears while those of x are replaced with n. To distinguish these two kinds of variable occurrences, we call the x in the function header a binding occurrence and those in the function’s body the bound occurrences. We also say that the binding occurrence of x binds all occurrences of x in the body of f. Indeed, people who study programming languages even have a name for the region where a binding occurrence works, namely, its lexical scope.

The definitions of f and g bind two more names: f and g. Their scope is called top-level scope because we think of scopes as nested (see below).

The word free occurrence applies to a variable without any binding occurrence. It is a name without definition, i.e., neither the language nor its libraries nor the program associates it with some value. For example, if you were to put the above program into a definitions area by itself and run it, entering f, g, and h at the prompt of the interactions would show that the first two are defined and the last one is not:
> f


> g


> h

h: this variable is not defined

The description of lexical scope suggests a pictorial representation of f’s definition: DrRacket’s “Check Syntax” functionality draws these diagrams.

Here is an arrow diagram for top-level scope:

Note that the scope of f includes all definitions above and below its definition. The bullet over the first occurrence indicates that it is a binding occurrence. The arrows from the binding occurrence to the bound occurrences suggest the flow of values. When the value of a binding occurrence becomes known, the bound occurrences receive their values from there.

Along similar lines, these diagrams also explain how renaming works. If you wish to rename a function parameter, you search for all bound occurrences in scope and replace them. For example, renaming f’s x to y in the program above means
(define (f x) (+ (* x x) 25))
(define (g x) (+ (f (+ x 1)) (f (- x 1))))
changes only two occurrences of x:
(define (f y) (+ (* y y) 25))
(define (g x) (+ (f (+ x 1)) (f (- x 1))))

Exercise 287. Here is a simple ISL+ program:
(define (p1 x y)
  (+ (* x y)
     (+ (* 2 x)
        (+ (* 2 y) 22))))
(define (p2 x)
  (+ (* 55 x) (+ x 11)))
(define (p3 x)
  (+ (p1 x 0)
     (+ (p1 x 1) (p2 x))))
Draw arrows from p1’s x parameter to all its bound occurrences. Draw arrows from p1 to all bound occurrences of p1. Check the results with DrRacket “Check Syntax” functionality. image

In contrast to top-level function definitions, the scope of the definitions in a local is limited. Specifically, the scope of local definitions is the local expression. Consider the definition of an auxiliary function f in a local expression. It binds all occurrences within the local expression but none that occur outside:

The two occurrences outside of local are not bound by the local definition of f. As always, the parameters of a function definition, local or not, are only bound in the function’s body.

Since the scope of a function name or a function parameter is a textual region, people also draw box diagrams to indicate scope. More precisely, for parameters a box is drawn around the body of a function:

In the case of a local definition, the box is drawn around the entire local expression:

In this example, the box describes the scope of the definitions of f and g.

Using a box for a scope, we can also easily understand what it means to reuse the name of function inside a local expression:

The inner box describes the scope of the inner definition of f; the outer box is the scope of the outer definition of f. Accordingly, all occurrences of f in the inner box refer to the inner local; all those in the outer box refer to the definition in the outer local. In other words, the scope of the outer definition of f has a hole, namely, the inner box.

Holes can also occur in the scope of a parameter definition. Here is an example:

In this function, the parameter x is used twice: for f and g. The scope of the latter is nested in the scope of the former and is thus a hole for the scope of the outer use of x.

In general, if the same name occurs more than once in a function, the boxes that describe the corresponding scopes never overlap. In some cases the boxes are nested within each other, which gives rise to holes. Still, the picture is always that of a hierarchy of smaller and smaller nested boxes.

Exercise 288. Here is a simple ISL+ function:
; [List-of X] -> [List-of X]
; creates a version of the given list that is sorted in descending order
(define (insertion-sort alon)
  (local ((define (sort alon)
              [(empty? alon) '()]
              [else (add (first alon) (sort (rest alon)))]))
          (define (add an alon)
              [(empty? alon) (list an)]
                  [(> an (first alon)) (cons an alon)]
                  [else (cons (first alon) (add an (rest alon)))])])))
    (sort alon)))
Draw a box around the scope of each binding occurrence of sort and alon. Then draw arrows from each occurrence of sort to the appropriate binding occurrence. Now repeat the exercise for the following variant:
(define (sort alon)
  (local ((define (sort alon)
              [(empty? alon) '()]
              [else (add (first alon) (sort (rest alon)))]))
          (define (add an alon)
              [(empty? alon) (list an)]
                  [(> an (first alon)) (cons an alon)]
                  [else (cons (first alon) (add an (rest alon)))])])))
    (sort alon)))
Do the two functions differ other than in name? image

Exercise 289. Recall that each occurrence of a variable receives its value from the corresponding binding occurrence. Consider the following definition:

(define x (cons 1 x))

Where is the shaded occurrence of x bound? Since the definition is a constant definition and not a function definition, we need to evaluate the right-hand side if we wish to work with this function. What should be the value of the right-hand side according to our rules? image

As discussed in Computing with lambda, a lambda expression is just a short-hand for a local expression, i.e.,

(lambda (x-1 ...  x-n) exp)

is short for
(local ((define (a-new-name x-1 ... x-n) exp))
  a-new-name) ; if a-new-name does not occur in exp.
The short-hand explanation suggests that

(lambda (x-1 ...  x-n) exp)

introduces x-1, ..., x-n as binding occurrences and that the scope of parameters is exp, e.g.,

Of course, if exp contains further binding constructs (say, a nested local expression), then the scope of the variables may have a hole.

Exercise 290. Draw arrows from the shaded occurrences of x to their binding occurrences in each of the following three lambda expressions:
  1. (lambda (x y)
      (+ x (* x y)))
  2. (lambda (x y)
      (+ x
         (local ((define x (* y y)))
           (+ (* 3 x)
              (/ 1 x)))))
  3. (lambda (x y)
      (+ x
         ((lambda (x)
            (+ (* 3 x)
               (/ 1 x)))
          (* y y))))
Also draw a box for the scope of each shaded x and holes in the scope as necessary. image

For Loops

Even though it never mentions the word, Abstraction introduces loops. Abstractly, a loop traverses compound data, processing one piece at a time. In the process, loops also synthesize data. For example, map traverses a list, applies a function to each item, and collects the results in a list. Similarly, build-list enumerates the sequence of predecessors of a natural number (from 0 to (- n 1)), maps each of these to some value, and also gathers the results in a list.

The loops of ISL+ differ from those in conventional languages in two ways. First, a conventional loop does not directly create new data; in contrast, abstractions such as map and build-list are all about computing new data from traversals. Second, conventional languages often provide only a fixed number of loops; an ISL+ programmer defines new loops as needed. Put differently, conventional languages view loops as syntactic constructs akin to local or cond, and their introduction requires a detailed explanation of their vocabulary, grammar, scope, and meaning.

Loops as syntactic constructs have two advantages over functional loops. On one hand, their shape tends to signal intentions more directly than a composition of functions. On the other hand, language implementations typically translate syntactic loops into faster commands for computers than functional loops. It is therefore common that even functional programming languages—with all their emphasis on functions and function compositions—provide syntactic loops.

Use the 2htdp/abstraction library for this section and the next one.

In this section, we introduce a few of Racket’s so-called for loops. The goal is to illustrate how to think about conventional loops as linguistic constructs and to indicate how programs built with abstractions may use loops instead. Figure 70 spells out the grammar of our selected for loops as an extension of BSL’s grammar from Intermezzo: BSL. Every loop is an expression and, like all compound constructs, is marked with a keyword. The latter is followed by a parenthesized sequence of so-called comprehension clauses and a single expression. The clauses introduce so-called loop variables, and the expression at the end is the loop body.

  expr = ...
  | (for/list (clause clause ...) expr)
  | (for*/list (clause clause ...) expr)
  | (for/and (clause clause ...) expr)
  | (for*/and (clause clause ...) expr)
  | (for/or (clause clause ...) expr)
  | (for*/or (clause clause ...) expr)
  | (for/sum (clause clause ...) expr)
  | (for*/sum (clause clause ...) expr)
  | (for/product (clause clause ...) expr)
  | (for*/product (clause clause ...) expr)
  | (for/string (clause clause ...) expr)
  | (for*/string (clause clause ...) expr)
  clause = (variable expr)

Figure 70: ISL+ extended with for loops

Even a cursory look at the grammar shows that the dozen looping constructs come in six pairs: a for and for* variant for each of list, and, or, sum, product, and string.Racket comes with many more loops than this, and its loops come with more functionality than those presented here. All for loops bind the variables of their clauses in the body; the for* variants also bind variables in the subsequent clauses. The following two, near-identical code snippets illustrate the difference between these two scoping rules:


The syntactic difference is that the left one uses for/list and the right one for*/list. In terms of scope, the two strongly differ as the arrows indicate. While both pieces introduce two loop variables—width and heightthe left one uses an externally defined variable for height’s initial value and the right one uses the first loop variable.

Semantically, a for/list expression evaluates the expressions in its clauses to generate sequences of values. If an expression in a clause evaluates to a
  • list, its items make up the sequence values;

  • natural number n, the sequence of values consists of the numbers 0, 1, ..., (- n 1);

  • string, its one-character strings are the sequence items.

Finally, for/list evaluates the loop body with the loop variables successively bound to the values of the sequences generated by the clause expressions, and it collects the values of its body into a list. Note that the evaluation of a for/list expression stops when the shortest sequence is exhausted.

Terminology Each evaluation of a loop body is called an iteration. Similarly, a loop is said to iterate over the values of its loop variables.

Based on this explanation, the following, first example generates the list from 0 to 9:
> (for/list ((i 10))

(list 0 1 2 3 4 5 6 7 8 9)

> (build-list 10 (lambda (i) i))

(list 0 1 2 3 4 5 6 7 8 9)

For comparison, the second expression shows how to create the same list with build-list and lambda. Here is a second example:
> (for/list ((i 2) (j '(a b)))
    (list i j))

(list (list 0 'a) (list 1 'b))

> (local ((define i-s (build-list 2 (lambda (i) i)))
          (define j-s '(a b)))
    (map list i-s j-s))

(list (list 0 'a) (list 1 'b))

Again, the second expression generates the same list using ISL+’s abstractions.

Let’s finish the introduction of for/list with a practical example:

Sample Problem: Design enumerate. The function consumes a list and produces a list of the same items paired with their relative index.

Stop! Design this function systematically using ISL+’s abstractions.

With for/list, this problem has a straightforward solution:
; [List-of X] -> [List-of [List N X]]
; pair each item in l with its index
(check-expect (enumerate '(a b c)) '((1 a) (2 b) (3 c)))
(define (enumerate l)
  (for/list ((item l) (ith (length l)))
    (list (+ ith 1) item)))
The function’s body uses for/list to iterate over the given list and a list of numbers from 0 to (- (length l) 1); the loop body combines the index with the list item.

In terms of semantics, for*/list differs from for/list in a subtle way. Unlike the latter, the former iterates over the sequences of values in a nested fashion. That is, a for*/list expression basically unfolds into a nest of loops:
> (for*/list ((i 2) (j '(a b)))
    (list i j))

(list (list 0 'a) (list 0 'b) (list 1 'a) (list 1 'b))

> (for/list ((i 2))
    (for/list ((j '(a b)))
       (list i j)))


 (list (list 0 'a) (list 0 'b))

 (list (list 1 'a) (list 1 'b)))

Nested loops, however, produce nested results; to collect those results into a single list, for*/list concatenates the nested lists using foldl and append.

Here is a problem that illustrates this use of for*/list in context:

Sample Problem: Design cross. The function consumes two lists, l1 and l2, and produces pairs of all items from l1 and l2.

Stop! Take a moment to design the function using existing abstractions.

If you designed cross properly, you worked through a table such as this one:










(list 'a 1)


(list 'b 1)


(list 'c 1)



(list 'a 2)


(list 'b 2)


(list 'c 2)

The first row displays l1 as given, while the left-most column shows l2. Each cell in the table corresponds to one of the pairs that cross must generate.

Since the purpose of for*/list is an enumeration of all such pairs, defining cross via for*/list is straightforward:
; [List-of X] [List-of Y] -> [List-of [List X Y]]
; generate all possible pairs of items from l1 and l2
(check-satisfied (cross '(a b c) '(1 2)) (lambda (c) (= (length c) 6)))
(define (cross l1 l2)
   (for*/list ([item1 l1][item2 l2])
      (list item1 item2)))
We use check-satisfied instead of check-expect because we do not wish to predict the exact order in which for*/list generates the pairs.

Note Figure 71 shows another in-context use of for*/list.We thank Mr. Mark Engelberg for suggesting this exhibition of expressive power. It displays a compact solution of the extended design problem covered by Word Games, the Heart of the Problem, namely,

given a word, create all possible re-arrangements of the letters in a list.

The extended exercise is a direct guide to the design of the main function and two auxiliaries, where the design of the latter requires the creation of two more helper functions. In contrast, figure 71 uses the combined power of for*/list and an unusual form of recursion to define the same program as a single, five-line function definition. The code is listed here merely to exhibit the power of these abstractions; for the yet-unknown design method, see Generative Recursion and especially exercise 451. End

; [List-of X] -> [List-of [List-of X]]
; creates a list of all rearrangements of the items in w
(define (arrangements w)
    [(empty? w) '(())]
    [else (for*/list ([item w]
                       (arrangements (remove item w))])
            (cons item arrangement-without-item))]))
; test:
; [List-of X] -> Boolean
(define (all-words-from-rat? w)
  (and (member? (explode "rat") w)
       (member? (explode "art") w)
       (member? (explode "tar") w)))
(check-satisfied (arrangements '("r" "a" "t")) all-words-from-rat?)

Figure 71: A compact definition of arrangements with for*/list

At this point, the difference between for/list and for*/list is pretty clear, but here is a pair of interactions that turns the difference in scope into different behavior:
> (define width 2)
> (for/list ([width 3][height width])
     (list width height))

(list (list 0 0) (list 1 1))

> (for*/list ([width 3][height width])
     (list width height))

(list (list 1 0) (list 2 0) (list 2 1))

To understand the first one, remember that for/list traverses the two sequences in parallel and stops when the shorter one is exhausted. Here, the two sequences are





















(list 0 0)


(list 1 1)

The first two rows show the values of the two loop variables, which change in tandem. The last row shows the result of each iteration, which explains the first result. Now contrast this situation with for*/list, which proceeds in a nested fashion:
















0, 1





(list 1 0)


(list 2 0), (list 2 1)

While the first row is like the one for for/list, the second one now displays sequences of numbers in its cells. The implicit nesting of for*/list means that each iteration with a specific value of width re-initializes height and thus creates a distinct sequence of height values. This explains why the first cell of height values is empty; after all, there are no natural numbers between 0 (inclusive) and 0 (exclusive). Finally, each nested for loop yields a sequences of pairs, which, as mentioned above, are collected into a single list of pairs.

It is also clear now that .../list signals that the loop creates a list by collecting the results in the order in which they are generated. By analogy, for and for* loops for
  • .../and apply an operation like and to all of the generated values:
    > (for/and ((i 10)) (> (- 9 i) 0))


    > (for/and ((i 10)) (if (>= i 0) i #false))


    For pragmatic reasons, the loops return the last generated value or #false if any of the results are #false.

  • .../or apply an operation like or to all of the generated values:
    > (for/or ((i 10)) (if (= (- 9 i) 0) i #false))


    > (for/or ((i 10)) (if (< i 0) i #false))


    These loops return the first value that is not #false unless all the values are #false.

  • .../sum add up the numbers that the iterations generate:
    > (for/sum ((c "abc")) (string->int c))


  • .../product multiply the numbers that the iterations generate
    > (for/product ((c "abc")) (+ (string->int c) 1))


  • .../string create Strings from the 1Strings that the iterations generate:
    > (define a (string->int "a"))
    > (for/string ((j 10)) (int->string (+ a j)))


Racket also comes with for/fold, which combines the results of the loop bodies via a program-defined operation; this loop is not included in the library.

Stop! It is an instructive exercise to re-formulate all of the above examples using the existing abstractions in ISL+. Doing so also indicates how to design functions with for loops instead of abstract functions. Hint Design and-map and or-map, which work like andmap and ormap, respectively, but return the appropriate non-#false values.

; N -> sequence?
; construct an infinite sequence of natural numbers starting at n
(define (in-naturals n) ...)
; N N N -> sequence?
; construct the finite sequence of natural numbers starting with
;   start
;   (+ start step)
;   (+ start step step)
;   ...
;  until the number exceeds end
(define (in-range start end step) ...)

Figure 72: Constructing sequences of natural numbers

Looping over numbers isn’t always a matter of enumerating 0 through (- n 1). Often programs need to step through non-sequential sequences of numbers; other times, an unlimited supply of numbers is needed. To accommodate this form of programming, Racket comes with functions that generate sequences, and figure 72 lists two that are provided in the abstraction library for ISL+.

With the first one, we can simplify the enumerate function a bit:
(define (enumerate.v2 l)
  (for/list ((item l) (ith (in-naturals 1)))
    (list ith item)))
Here in-naturals is used to generate the infinite sequence of natural numbers starting at 1; the for loop stops when l is exhausted. Note This version of enumerate is also a bit faster than the original because it traverses l only once.

With the second one, it is, for example, possible to step through the even numbers among the first n:
; N -> Number
; add the even numbers between 0 and n (exclusive)
(check-expect (sum-evens 2) 0)
(check-expect (sum-evens 4) 2)
(define (sum-evens n)
  (for/sum ([i (in-range 0 n 2)]) i))
Although this use may appear trivial, many problems originating in mathematics call for just such loops, which is precisely why concepts such as in-range are found in many programming languages.

Exercise 291. Use loops to define convert-euro, a function that converts a list of US$ amounts into a list of € amounts based on an exchange rate of €1.08 per US$. Compare with exercise 255. image

Exercise 292. Use loops to define a function that
  1. creates the list 0 ... (n - 1) for any natural number n;

  2. creates the list 1 ... nfor any natural number n;

  3. creates the list (list 1 1/10 1/100 ...) of n numbers for any natural number n;

  4. creates the list of the first n even numbers;

  5. creates a list of lists of 0 and 1 in a diagonal arrangement, e.g.,

    (equal? (diagonal 3)
              (list 1 0 0)
              (list 0 1 0)
              (list 0 0 1)))
Finally, use loops to define tabulate from exercise 238. See exercise 258. image

Exercise 293. Define find-name. The function consumes a name and a list of names. It retrieves the first name on the latter that is equal to, or an extension of, the former.

Define a function that ensures that no name on some list of names exceeds some given width. Compare with exercise 259. image

Pattern Matching

When we design a function for a data definition with six clauses, we use a six-pronged cond expression.The 2htdp/abstraction library also supports the definition of algebraic data types with define-type and their deconstruction using type-case. The latter is like match but also ensures that the conditional has the correct number and kinds of clauses. When we formulate one of the cond clauses, we use a predicate to determine to determine whether this clause should process the given value and, if so, selectors to deconstruct any compound values. The first three parts of this book explain this idea over and over again, and many of its code snippets exhibit just this pattern.

Repeated patterns call for abstraction. While Abstraction explains how programmers can create some of these abstractions, the predicate-selector pattern can be addressed only by a language designer. In particular the designers of functional programming languages have recognized the need for abstracting the repetitive uses of predicates and selectors. These languages therefore provide pattern matching as a linguistic construct that combines and simplifies these cond clauses.

This section presents Racket’s pattern matcher as provided by the 2htdp/abstraction library. Figure 73 displays the grammatical extension for match. Clearly, match is a syntactically complex construct. While its basic outline resembles that of cond, the conditions are replaced by patterns, which come with their own vocabulary and grammatical rule.

  expr = ...
  | (match expr [pattern expr] ...)
  pattern = variable
  | literal-constant
  | (cons pattern pattern)
  | (structure-name pattern ...)
  | (? predicate-name)

Figure 73: ISL+ match expressions

Roughly speaking,
(match expr
  [pattern1 expr1]
  [pattern2 expr2]
proceeds like a cond expression in that it evaluates expr and sequentially tries to match its result with pattern1, pattern2, ... until it succeeds with patterni. At that point, it determines the value of expri, which is also the result of the entire match expression.

The key difference is that match, unlike cond, introduces a new scope, which is best illustrated with a screen-shot from DrRacket:

As the image shows, each pattern clause of this function binds variables. Furthermore, the scope of a variable is the body of the clause, so even if two patterns introduce the same variable binding—as it is the case in the above code snippet—their bindings cannot interfere with each other.

Syntactically, a pattern resembles a piece of nested, structural data whose leafs are literal constants, variables, or predicate patterns of the shape (? predicate-name). In the latter, predicate-name must refer to a predicate function in scope, that is, a function that consumes one value and produces a Boolean.

Semantically, a pattern is matched to a value v. If the pattern is
  • a literal-constant, it matches only that literal constant

    > (match 4
        ['four  1]
        ["four" 2]
        [#true  3]
        [4      "hello world"])

    "hello world"

  • a variable, it matches any value, and it is associated with this value during the evaluation of the body of the corresponding match clause

    > (match 2
        [3 "one"]
        [x (+ x 3)])


    Since 2 does not equal the first pattern, which is the literal constant 3, match matches 2 with the second pattern, which is a plain variable and thus matches any value. Hence, match picks the second clause and evaluates its body, with x standing for 2.

  • (cons pattern1 pattern2), it matches only an instance of cons, assuming its first field matches pattern_1 and its rest field matches pattern_2

    > (match (cons 1 '())
        [(cons 1 tail) tail]
        [(cons head tail) head])


    > (match (cons 2 '())
        [(cons 1 tail) tail]
        [(cons head tail) head])


    These interactions show how match first deconstructs cons and then uses literal constants and variables for the leafs of the given list.

  • (structure-name pattern1 ... patternn), it matches only a structure-name structure, assuming its field values match pattern1, ..., patternn

    > (define p (make-posn 3 4))
    > (match p
        [(posn x y) (sqrt (+ (sqr x) (sqr y)))])


    Obviously, matching an instance of posn with a pattern is just like matching a cons pattern. Note though, how the pattern uses posn for the pattern, not the name of the constructor.

    Matching also works for programmer-introduced structure type definitions:
    > (define-struct phone [area switch four])
    > (match (make-phone 713 664 9993)
        [(phone x y z) (+ x y z)])


    Again, the pattern uses the name of the structure, phone, not that of the constructor.

    Finally, matching also works across several layers of data constructions:
    > (match (cons (make-phone 713 664 9993) '())
        [(cons (phone area-code 664 9993) tail) area-code])


    This match expression extracts the area code from a phone number in a list if the switch code is 664 and the last four digits are 9993.

  • (? predicate-name), it matches only when (predicate-name v) produces #true

    > (match (cons 1 '())
        [(cons (? symbol?) tail) tail]
        [(cons head tail) head])


    This expression produces 1, the result of the second clause, because 1 is not a symbol.

Stop! Experiment with match before you read on.

At this point, it is time to demonstrate the usefulness of match in context:

Sample Problem: Design the function last-item, which retrieves the last item on a non-empty list. Recall that non-empty lists are defined as follows:
; A [Non-empty-list X] is one of:
;  (cons X '())
;  (cons X [Non-empty-list X])

Stop! Arbitrarily Large Data deals with this problem. Look up the solution.

With match, a designer can eliminate three selectors and two predicates from the solution using cond:
; [Non-empty-list X] -> X
; retrieve the last item of ne-l
(check-expect (last-item '(a b c)) 'c)
(check-error (last-item '()))
(define (last-item ne-l)
   (match ne-l
     [(cons lst '()) lst]
     [(cons fst rst) (last-item rst)]))
Instead of predicates and selectors, this solution uses patterns that are just like those found in the data definition. For each self-reference and occurrence of the set parameter in the data definition, the patterns use program-level variables. The bodies of the match clauses no longer extract the relevant parts from the list with selectors but simply refer to these names. As before, the function recurs on the rest field of the given cons because the data definition refers to itself in this position. In the base case, the answer is lst, the variable that stands for the last item on the list.

Let’s take a look at a second problem from Arbitrarily Large Data:

Sample Problem: Design the function depth, which measures the number of layers surrounding Russian doll. Recall the corresponding data definition:
(define-struct layer [color doll])
; An RD (russian doll) is one of:
;  "wooden doll"
;  (make-layer String RD)

Stop! Look up the solution and try to modify it to use match.

Here is a definition of depth using match:
; RD -> N
; how many dolls are a part of an-rd
(check-expect (depth (make-layer "red" "wooden doll")) 1)
(define (depth a-doll)
  (match a-doll
    ["wooden doll" 0]
    [(layer c inside) (+ (depth inside) 1)]))
While the pattern in the function’s first match clause looks for the literal string "wooden doll", the second one matches any layer structure, associating c with the value in the color field and inside with the value in the doll field. In short, match again simplifies the conditional expression needed by eliminating predicates and selectors.

The final problem is an excerpt from a generalized version of our UFO game:

Sample Problem: Use match to design the function move-right. The latter consumes a list of Posns, which represent the positions of objects on a canvas, plus a number. It adds the latter to each x-coordinate, which represents an rightward movement of these objects.

Stop! Look up the function tock in figure 43, which solves a variant of this sample problem.

Our solution to the problem combines the abstract map function introduced in the preceding part of the book with match:
; [List-of Posn] -> [List-of Posn]
; moves each object right by delta-x pixels
(check-expect (move-right (list (make-posn 1 1) (make-posn 10 14)) 3)
              (list (make-posn 4 1) (make-posn 13 14)))
(define (move-right lop delta-x)
  (map (lambda (p)
         (match p
           [(posn x y) (make-posn (+ x delta-x) y)]))
You may wish to define the version that uses cond and the selectors and the one that uses match only—both without using map. We leave it to you to judge which version is the most readable one.

Exercise 294. Use match to design the function replace, which substitutes the area code 713 with 281 in a list of phone records. For a structure type definition of phone records, see above. Formulate a suitable data definition first. If you are stuck, look up your solution for exercise 170. image

Exercise 295. Figure 47 displays a function that determines the number of words per nested list in a list of list of strings. Using the notation from Abstraction, the function’s signature, purpose and header can be formulated like this:
; [List-of [List-of String]] -> [List-of Number]
; determines the number of words on each line
(define (words-on-line lls)
Define the function using match. image