1. 6.001 SICP 1 6.001 SICP – October 20 6001-Introduction Trevor Darrell trevor@csail.mit.edu 32-D512 Office Hour: W 11 6.001 web page: http://sicp.csail.mit.edu/ section web page: http://www.csail.mit.edu/~trevor/6001/ sets adts tables – assoc tables – hash

2. 6.001 SICP 2 Set data abstractions Represent a set as a list of its elements in which no element appears more than once. (define (element-of-set? x set) (cond ((null? set) false) ((equal? x (car set)) true) (else (element-of-set? x (cdr set))))) (define (element-of-set? x set) (not (null? (filter (lambda (e) (equal? e x)) set)))

3. 6.001 SICP 3 Adjoin-set If the object to be adjoined is already in the set, we just return the set. Otherwise, cons it on! (define (adjoin-set x set) (if (element-of-set? x set) set (cons x set))) We can define union-set in one line using adjoin-set: (define (union-set set1 set2) (accumulate adjoin-set set1 set2))

4. 6.001 SICP 4 Intersection-set Assume we know how to form the intersection of set2 and the cdr of set1… (define (intersection-set set1 set2) (cond ((or (null? set1) (null? set2)) '()) ((element-of-set? (car set1) set2) (cons (car set1) (intersection-set (cdr set1) set2))) (else (intersection-set (cdr set1) set2)))) Alternatively, use filter: (define (intersection-set set1 set2) (filter (lambda (x) (element-in-set? x set2)) set1)))

5. 6.001 SICP 5 Complexity Element-of-set? may have to scan the entire set. Intersection-set does an element-of-set? check for each element of set1 The complexity of intersection/union is: O(n^2) for two sets of size n

6. 6.001 SICP 6 Sets as ordered lists Representation optimized for search: Set elements are listed in increasing order. In checking for the presence of an item, we no longer have to scan the entire set: (define (element-of-set? x set) (cond ((null? set) false) ((= x (car set)) true) ((< x (car set)) false) (else (element-of-set? x (cdr set))))) Thus, the average number of steps required will be about n/2. This is still O(n) growth, but it does save us, on the average, a factor of 2.

7. 6.001 SICP 7 Ordered intersection-set But this really helps with intersection-set. Unordered representation this operation required O(n^2) steps, because we performed a complete scan of set2 for each element of set1. But with the ordered representation, we can use a linear method….how?

8. 6.001 SICP 8 Ordered intersection-set (define (intersection-set set1 set2)

9. 6.001 SICP 9 Ordered intersection-set (define (intersection-set set1 set2) (if (or (null? set1) (null? set2)) '() (let ((x1 (car set1)) (x2 (car set2))) (cond ((= x1 x2)

10. 6.001 SICP 10 Ordered intersection-set (define (intersection-set set1 set2) (if (or (null? set1) (null? set2)) '() (let ((x1 (car set1)) (x2 (car set2))) (cond ((= x1 x2) (cons x1 (intersection-set (cdr set1) (cdr set2)))) ((< x1 x2)

11. 6.001 SICP 11 Ordered intersection-set (define (intersection-set set1 set2) (if (or (null? set1) (null? set2)) '() (let ((x1 (car set1)) (x2 (car set2))) (cond ((= x1 x2) (cons x1 (intersection-set (cdr set1) (cdr set2)))) ((< x1 x2) (intersection-set (cdr set1) set2)) ((< x2 x1) (intersection-set set1 (cdr set2)))))))

12. 6.001 SICP 12 Traditional LISP structure: association list A list where each element is a list of the key and value. 15 x 20 y x: 15 y: 20 Represent the table as the alist: ((x 15) (y 20))

13. 6.001 SICP 13 Alist operation: find-assoc (define (find-assoc key alist) (cond ((null? alist) #f) ((equal? key (caar alist)) (cadar alist)) (else (find-assoc key (cdr alist))))) (define a1 '((x 15) (y 20))) (find-assoc 'y a1) ==> 20

14. 6.001 SICP 14 Alist operation: add-assoc (define (add-assoc key val alist) (cons (list key val) alist)) (define a2 (add-assoc 'y 10 a1)) a2 ==> ((y 10) (x 15) (y 20)) (find-assoc 'y a2) ==> 10

15. 6.001 SICP 15 Alist operation: find-assoc (find-assoc ‘bar ‘((foo 6) (jane 3)) ==> #f (find-assoc ‘bar ‘((foo 6) (bar #f) (ed 4)) ==> #f

16. 6.001 SICP 16 find-assoc-entry (define (find-assoc-entry key alist) (cond ((null? alist) #f) ((equal? key (caar alist)) ??? ) (else (find-assoc-entry key (cdr alist))))) (find-assoc-entry ‘bar ‘((foo 6) (bar #f) (ed 4)) ==> (bar #f)

17. 6.001 SICP 17 find-assoc-entry (define (find-assoc-entry key alist) (cond ((null? alist) #f) ((equal? key (caar alist)) (car alist)) (else (find-assoc-entry key (cdr alist))))) (find-assoc-entry ‘bar ‘((foo 6) (bar #f) (ed 4)) ==> (bar #f)

18. 6.001 SICP 18 Alists are not an abstract data type Missing a constructor: Use quote or list to construct (define a1 '((x 15) (y 20))) There is no abstraction barrier: Definition in scheme language manual: "An alist is a list of pairs, each of which is called an association. The car of an association is called the key." Therefore, the implementation is exposed. User may operate on alists using list operations. (filter (lambda (a) ( ((x 15))

19. 6.001 SICP 19 Why do we care that Alists are not an ADT? Modularity is essential for software engineering Build a program by sticking modules together Can change one module without affecting the rest Alists have poor modularity Programs may use list ops like filter and map on alists These ops will fail if the implementation of alists change Must change whole program if you want a different table To achieve modularity, hide information Hide the fact that the table is implemented as a list Do not allow rest of program to use list operations ADT techniques exist in order to do this

20. 6.001 SICP 20 Table: a set of bindings binding: a pairing of a key and a value Abstract interface to a table: make create a new table put! key value insert a new binding replaces any previous binding of that key get key look up the key, return the corresponding value This definition IS the table abstract data type Code shown later is an implementation of the ADT

21. 6.001 SICP 21 Table1: Table ADT implemented as an Alist (define table1-tag 'table1) (define (make-table1) (cons table1-tag nil)) (define (table1-get tbl key) (find-assoc key (cdr tbl))) (define (table1-put! tbl key val) (set-cdr! tbl (add-assoc key val (cdr tbl))))

22. 6.001 SICP 22 Examples of using tables Fred John Bill People 34 48 Age 2000 1999 1998 Age Job Pay 34

23. 6.001 SICP 23 ADT and tags standard pattern for an ADT with tagged data 1.a variable in the ADT implementation stores the tag 2.attach the tag in the constructor 3.write a predicate that checks the tag 4.operations strip the tags, operate, attach the tag again Use tagged data everywhere (including return values) Using tagged data is only defensive programming if you check the tags and don't use the else branch Traditionally, ADT operations / accessors don't check tags, but paranoia is fine here.

24. 6.001 SICP 24 Table2: Table ADT implemented as hash table (define t2-tag 'table2) (define (make-table2 size hashfunc) (let ((buckets (make-vector size nil))) (list t2-tag size hashfunc buckets))) (define (size-of tbl) (cadr tbl)) (define (hashfunc-of tbl) (caddr tbl)) (define (buckets-of tbl) (cadddr tbl))

25. 6.001 SICP 25 get in table2 (define (table2-get tbl key) (let ((index ((hashfunc-of tbl) key (size-of tbl)))) (find-assoc key (vector-ref (buckets-of tbl) index)))) (define (table2-put! tbl key val) (let ((index ((hashfunc-of tbl) key (size-of tbl))) (buckets (buckets-of tbl))) (vector-set! buckets index (add-assoc key val (vector-ref buckets index)))))

26. 6.001 SICP 26 Table2 example (define tt2 (make-table2 4 hash-a-point)) (table2-put! tt2 (make-point 5 5) 20) (table2-put! tt2 (make-point 5 7) 15) tt2 table2 4 vector 15 point 5,7 20 point 5,5 (table2-get tt2 (make-point 5 5)) (table2-put! tt2 (make-point 5 5) 20) (table2-put! tt2 (make-point 5 7) 15) (table2-get tt2 (make-point 5 5)) (define t2-tag 'table2) (define (make-table2 size hashfunc) (let ((buckets (make-vector size nil))) (list t2-tag size hashfunc buckets))) (define (size-of tbl) (cadr tbl)) (define (hashfunc-of tbl) (caddr tbl)) (define (buckets-of tbl) (cadddr tbl)) (define (table2-get tbl key) (let ((index ((hashfunc-of tbl) key (size-of tbl)))) (find-assoc key (vector-ref (buckets-of tbl) index)))) (define (table2-put! tbl key val) (let ((index ((hashfunc-of tbl) key (size-of tbl))) (buckets (buckets-of tbl))) (vector-set! buckets index (add-assoc key val (vector-ref buckets index)))))

27. 6.001 SICP 27 Table2 example (define tt2 (make-table2 4 hash-a-point)) (table2-put! tt2 (make-point 5 5) 20) (table2-put! tt2 (make-point 5 7) 15) 15 point 5,7 20 point 5,5 (table2-get tt2 (make-point 5 5)) tt2 table2 4 vector

28. 6.001 SICP 28 Table2 example (define tt2 (make-table2 4 hash-a-point)) (table2-put! tt2 (make-point 5 5) 20) (table2-put! tt2 (make-point 5 7) 15) 15 point 5,7 20 point 5,5 (table2-get tt2 (make-point 5 5)) tt2 table2 4 vector

29. 6.001 SICP 29 Table2 example (define tt2 (make-table2 4 hash-a-point)) (table2-put! tt2 (make-point 5 5) 20) (table2-put! tt2 (make-point 5 7) 15) 15 point 5,7 (table2-get tt2 (make-point 5 5)) tt2 table2 4 vector 20 point 5,5

30. 6.001 SICP 30 Table2 example (define tt2 (make-table2 4 hash-a-point)) (table2-put! tt2 (make-point 5 5) 20) (table2-put! tt2 (make-point 5 7) 15) 15 point 5,7 (table2-get tt2 (make-point 5 5)) tt2 table2 4 vector 20 point 5,5

31. 6.001 SICP 31 Table2 example (define tt2 (make-table2 4 hash-a-point)) (table2-put! tt2 (make-point 5 5) 20) (table2-put! tt2 (make-point 5 7) 15) (table2-get tt2 (make-point 5 5)) 20 point 5,5 tt2 table2 4 vector 15 point 5,7

32. 6.001 SICP 32 Table2 example (define tt2 (make-table2 4 hash-a-point)) (table2-put! tt2 (make-point 5 5) 20) (table2-put! tt2 (make-point 5 7) 15) (table2-get tt2 (make-point 5 5)) 15 point 5,7 20 point 5,5 tt2 table2 4 vector

33. 6.001 SICP 33 Table2 example (define tt2 (make-table2 4 hash-a-point)) (table2-put! tt2 (make-point 5 5) 20) (table2-put! tt2 (make-point 5 7) 15) tt2 table2 4 vector 15 point 5,7 20 point 5,5 (table2-get tt2 (make-point 5 5))

34. 6.001 SICP 34 Hash functions What makes a good hash function? not too many indicies indicies are evenly distributed hash function is easy to compute E.g., how to hash products at Wal-mart? price? price/100? price modulo 100? SKU?