Clojure Lisp Reloaded Versions of LISP n Lisp

Clojure “Lisp Reloaded”

Versions of LISP n Lisp is an old language with many variants n n n LISP is an acronym for List Processing language Lisp is alive and well today Most modern versions are based on Common Lisp Scheme is one of the major variants Racket is the latest version of Scheme Clojure is the latest in a long line of dialects n Clojure uses a new approach to concurrency, Software Transactional Memory (STM) 2

Lisp syntax vs. Clojure syntax 3

Basic data types I n Numbers: Clojure has integers, floating point numbers, and ratios n n Strings are enclosed in double quotes, and can contain escaped characters n n Within a string, n represents a newline, as usual Character literals are written as c to indicate the character c n n Integers include decimal integers (255), octals numbers (012), hexadecimal numbers (0 xff) and radix numbers (2 r 1111), where the radix is an integer between 2 and 36, inclusive Floating point numbers include standard notation (3. 1416) and scientific notation(1. 35 e-12). Ratios are “fractions” such as 1/3; they are not subject to roundoff error Since n represents the character n, newline is used to represent a newline Keywords are not reserved words, but are like “atoms” in Prolog or Erlang; a keyword begins with a colon (: foo) and stands for itself 4

Basic data types II n A list is a sequence of values enclosed in parentheses, for example, (one 2 "buckle my shoe") n Elements of the list are separated by whitespace or commas n A list represents either data or a function call, depending on context n A vector is a sequence of values enclosed in brackets, for example, [one 2 "buckle my shoe”] n n A map or hash is a sequence of key/value pairs, enclosed in braces, for example, {: ace 1, : deuce 2, "trey" 3} n n n Elements of the vector are separated by whitespace or commas Elements are separated by whitespace or commas It is helpful to use commas between key/value pairs A set is a pound sign (#) followed sequence of values enclosed in braces, for example #{a b c} 5

Functions n The syntax to define a named function is: (defn function_name docstring? [arguments] n n n The docstring after the function name is optional The value of the function is the value of the last expression evaluated The syntax of a function call is (function_name arguments) n n expressions) Notice that the name of the function being called is the first thing inside the parentheses The syntax of an anonymous function, or lambda, is (fn [arguments] expressions) n n This syntax can be used in place of a function name Example: ((fn [lst] (first lst)) my-list) is equivalent to (first my-list) Alternative syntax: #(expressions), where %1 (or just %), %2, %3, … stand for parameters Since functions are first-class values, this syntax is convenient for creating a function to be used as an argument to another function 6

Sequences n n Lists, vectors, maps, and sets are all “sequences, ” and many of the same operations apply (first seq) returns the first element of seq n n (rest seq) returns the seq without its first element n n n If the seq is empty, the empty list, (), is returned (cons value seq) returns a sequence of the same type with the value inserted as the new first element Some sequences may be lazy; the elements of the sequence are not actually computed until they are used n This allows Clojure to have infinite sequences, for example, “all positive integers” 7

Fast and slow operations n Sequence is an interface roughly analogous to Java’s Iterator: n n Clojure. org describes sequences as “logical lists” n n n It is a small interface (a couple of methods) that can be applied to a large range of data types Many (most? ) of the common Clojure operations are defined on sequences For example, map and reduce work across almost all data types. The primary sequence operations are first, rest and cons Everything else is built around these operations. The set of cheap (constant time) operations are the same as for lists: Add/remove/access the first element or step to the next element Operations like nth and last are expensive (O(n) time) since they must traverse the entire list. count -- getting the length of a sequence -- is also expensive. 8

Sequence operations n Sequence operations can be used on almost any compound type n n n Collections: Vector, list, map, set, struct Strings (as char sequence) Java arrays, collections, sets, maps, iterator, enumerable IO streams Producer functions: Many lazy seqs do not get their values from a backing data set, but from a function that calculates the value when asked for. Some collections can be used directly as sequences, others need to be converted n The rules are somewhat different for different kinds of collections, so the simplest general strategy is to n n n convert to sequence, using seq do the processing if you need a particular non-sequence type (like vector or array), convert the return value back 9

Predicates n n A predicate (in any computer language) is a function that returns either “true” or “false” In Clojure, n n n “false” is one of the atoms false or nil “true” is the atom true In addition, anything that isn’t false or nil is considered to be true n n In particular, the empty list, (), is “true” Hence, a predicate returns either false, nil, or any other value to mean “true” n Predicates often return “true” values other than true, especially if the returned value might be useful 10

Function calls and data n A function call is written as a list n n n Example: (println "Hello" "World") n n The first element is the name of the function Remaining elements are the arguments Calls function println with arguments "Hello" and "World" The function prints Hello World and returns nil Data is written as atoms or lists Example: (println "Hello" "World") is a list of three elements 11

Quoting n n Is (f a b) a call to f, or is it just data? Lists and atoms must be quoted n n n x indicates the variable x, while (quote x) is just the atom x (quote (f a b)) is the list (f a b) n n Exceptions include nil, true, and nil, which do not need to be quoted quote is not a function, but a special form The special form gets the unevaluated arguments, and is control of when or whether to evaluate them Special forms, unlike functions, are not first-class values '(f a b) is a shorthand way of doing the same thing n n There is just one single quote at the beginning It quotes one expression 12

Examples n n (defn my-first [x] (first x)) user=> (my-first '(1 2 3)) 1 defn my-second [lst] (first (rest lst))) user=> (my-second '(a b c)) b n (defn my-third "This is a doc comment" [lst] (first (rest lst))) ) n user=> (my-third "Whatever") a n user=> (doc my-third) ------------user/my-third ([lst]) This is a doc comment nil 13

Arithmetic n n + * / n n n n n Returns the sum of its arguments; (+) returns 0 Subtracts the rest of the numbers from the first number Returns the product of its arguments; (*) returns 1. Divides the rest of the numbers into the first number If operands are integers, result is a ratio If only one number, returns its inverse quot Returns the quot[ient] of integer division of the first number by the rest of the numbers remainder of dividing the first number by the second mod Modulus of first and second number; truncates toward negative infinity inc Returns a number one greater than its argument dec Returns a number one less than its argument max Returns the largest of its arguments min Returns the smallest of its arguments 14

Numeric comparisons n = n n n n Returns true if all arguments are equal This is a true equality test, not an identity test Collections are compared in a type-independent manner; for example, (= '(1 2 3) [1 2 3]) returns true == Returns true if numeric arguments all have the same value, otherwise false not= Same as (not (= obj 1 obj 2)) < Returns true if numeric arguments are in monotonically increasing order > Returns true if numeric arguments are in monotonically decreasing order, otherwise false <= Returns true if numeric arguments are in monotonically non-decreasing order, otherwise false >= Returns true if numeric arguments are in monotonically non-increasing order, otherwise false 15

Conditional execution n if is a special form that takes exactly three arguments: n n n A predicate An expression to evaluate if the predicate is true An expression to evaluate if the predicate is false Syntax: (if Example: predicate true-branch false-branch) (defn sum [lst] (if (= lst ()) 0 (+ (first lst) (sum (rest lst))) ) ) (defn average [lst] (/ (sum lst) (count lst)) ) 16

The problem with if n if expressions nest rather awkwardly n n (defn collatz [n] (println n) (if (= n 1) 1 (if (= (rem n 2) 0) (collatz (/ n 2)) (collatz (inc (* 3 n))) ) In most cases, the more general cond is preferred 17

cond n cond implements the if. . . then. . . elseif. . . then. . . n control structure Like if, cond is a special form, not a function n That is, the arguments to cond are not evaluated before cond is called; rather, cond evaluates the arguments as it pleases 18

Syntax of the cond n n The syntax of the cond special form is: (condition result. . . ) Example: n n (defn pos-neg-or-zero "Determines whether n is positive, negative, or zero" [n] (cond (< n 0) "negative" (> n 0) "positive" : else "zero" ) ) The last condition, : else, is true, and reads better than the atom true 19

Rules for Recursion n n Handle the base (“simplest”) cases first Recur only with a “simpler” case n n n “Simpler” = more like the base case Don’t alter global variables Don’t look down into the recursion 20

Example: member n As an example we define member, to test membership in a list n n (defn member [a lat] (cond (empty? lat) false (= a (first lat)) true : else (member a (rest lat)) ) ) user=> (member : b '(: a : b : c)) true 21

Guidelines for recursive Clojure functions n n Unless the function is trivial, use a cond Handle the base case(s) first n n Avoid having more than one base case The base case is often testing for an empty list Do something with the first and recur with the rest Use tail recursion wherever possible n (To be explained later) 22

Example: union n (defn union [set 1 set 2] (cond (empty? set 1) set 2 (member (first set 1) set 2) (union (rest set 1) set 2) : else (cons (first set 1) (union (rest set 1) set 2)) ) ) This example uses the previously defined member function It appears that, if function A calls function B, Clojure requires function B to be defined before defining function A n n Of all the languages I have used, only C and its variants make this requirement Unless I’m missing something, this is a real step backward for Lisp dialects 23

Tests n The following can be used with any type of argument n nil? identical? Returns true if its argument is nil, false otherwise. Tests if its two arguments are the same object The following will throw an exception if given a non-numeric argument n n n zero? pos? neg? even? odd? Returns true if its argument is zero, else false Returns true if its argument is greater than zero Returns true if its argument is less than zero Returns true if its argument is even Returns true if its argument is odd 24

Type tests n coll? n seq? n vector? n list? n map? n set? Returns true if x implements IPersistent. Collection Return true if x implements ISeq Return true if x implements IPersistent. Vector Returns true if x implements IPersistent. List Return true if x implements IPersistent. Map Returns true if x implements IPersistent. Set 25

Content Tests n n n n (contains? map key) n Returns true if key is present in the map (distinct? args) n Returns true if no two of the arguments are = (empty? collection) n Returns true if collection has no items - same as (not (seq coll)) n Clojure programmers use the idiom (seq x) rather than (not (empty? x)) (every? predicate collection) n Returns true if (predicate x) is true for every x in collection (not-every? predicate collection) n Returns false if (predicate x) is true for every x in collection (not-any? predicate collection) n Returns false if (predicate x) is true for any x in collection (some predicate collection) n Returns the first value in the collection for which the predicate is true 26

I/O n n n *in* *out* *err* print, pr printf println prn newline read-line slurp spit A java. io. Reader object representing standard input A java. io. Writer object representing standard output A java. io. Writer object representing standard error Prints the object(s) to the *out* output stream, separated by spaces Prints formatted output, according to a format Same as print followed by newline Same as pr followed by newline. Observes *flush-on-newline* Writes a newline to the output stream *out* Reads the next line from stream *in* Reads the file into a string and returns it Opposite of slurp. Opens file with writer, writes content, then closes it 27

Dealing with side effects n Functions compute values; but printing is a side effect n n The value returned by println is nil—not too useful You can put multiple expressions in a defn n n The result of calling the function is the result of evaluating the last expression (defn five [] (println "Just five") 5 ) (five) Just five 5 You can also use do; the result is the value of the last expression n n (do (println "one" "two") (+ 1 2) ) one two 3 28

REPL commands n n n n n (load-filename) loads a file containing Clojure functions *1 bound in a repl thread to the most recent value printed *2 bound in a repl thread to the second most recent value printed *3 bound in a repl thread to the third most recent value printed *e bound in a repl thread to the most recent exception caught by the repl *print-dup* When set to logical true, objects will be printed in a way that preserves their type when read in later *print-length* controls how many items of each collection the print *print-level* controls how many levels deep the printer will print *print-meta* If set to logical true, when printing an object, its metadata will also be printed in a form that can be read back by the reader *print-readably* When set to logical false, strings and characters will be printed with non-alphanumeric characters converted to the appropriate escape sequences 29

Dealing with the REPL n The last time I used Clojure’s REPL, it seemed to be very buggy n I got a lot of errors along these lines: java. lang. Exception: Unable to resolve symbol: union in this context (NO_SOURCE_FILE: 43) n n Here are two workarounds that seem to solve most such problems: n n (It’s a Java exception because Clojure is compiled to the JVM) Start a new Clojure shell, or Just copy the function definition and paste it directly into the Clojure shell That was a year ago; the new version may be better You might also try out clooj, which is a very simple Clojure IDE n https: //github. com/arthuredelstein/clooj 30

The End 31