数据库系统基础教程第五章答案.doc

《数据库系统基础教程第五章答案.doc》由会员分享，可在线阅读，更多相关《数据库系统基础教程第五章答案.doc（16页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、数据库系统基础教程第五章答案Exercise 5.1.1As a set:speed2.662.101.422.803.202.202.001.863.06Average = 2.37As a bag:speed2.662.101.422.803.203.202.202.202.002.801.862.803.06Average = 2.48Exercise 5.1.2As a set:hd25080320200300160Average = 218As a bag:hd2502508025025032020025025030016016080Average = 215Exercise 5.1

2、.3aAs a set:bore15161418As a bag:bore1516141615151418Exercise 5.1.3bbore(Ships Classes)Exercise 5.1.4aFor bags:On the left-hand side:Given bags R and S where a tuple t appears n and m times respectively, the union of bags R and S will have tuple t appear n + m times. The further union of bag T with

3、the tuple t appearing o times will have tuple t appear n + m + o times in the final result. On the right-hand side:Given bags S and T where a tuple t appears m and o times respectively, the union of bags R and S will have tuple t appear m + o times. The further union of bag R with the tuple t appear

4、ing n times will have tuple t appear m + o + n times in the final result.For sets:This is a similar case when dealing with bags except the tuple t can only appear at most once in each set. The tuple t only appears in the result if all the sets have the tuple t. Otherwise, the tuple t will not appear

5、 in the result. Since we cannot have duplicates, the result only has at most one copy of the tuple t.Exercise 5.1.4bFor bags:On the left-hand side:Given bags R and S where a tuple t appears n and m times respectively, the intersection of bags R and S will have tuple t appear min( n, m ) times. The f

6、urther intersection of bag T with the tuple t appearing o times will produce tuple t min( o, min( n, m ) ) times in the final result. On the right-hand side:Given bags S and T where a tuple t appears m and o times respectively, the intersection of bags R and S will have tuple t appear min( m, o ) ti

7、mes. The further intersection of bag R with the tuple t appearing n times will produce tuple t min( n, min( m, o ) ) times in the final result.The intersection of bags R,S and T will yield a result where tuple t appears min( n,m,o ) times.For sets:This is a similar case when dealing with bags except

8、 the tuple t can only appear at most once in each set. The tuple t only appears in the result if all the sets have the tuple t. Otherwise, the tuple t will not appear in the result.Exercise 5.1.4cFor bags:On the left-hand side:Given that tuple r in R, which appears m times, can successfully join wit

9、h tuple s in S, which appears n times, we expect the result to contain mn copies. Also given that tuple t in T, which appears o times, can successfully join with the joined tuples of r and s, we expect the final result to have mno copies.On the right-hand side:Given that tuple s in S, which appears

10、n times, can successfully join with tuple t in T, which appears o times, we expect the result to contain no copies. Also given that tuple r in R, which appears m times, can successfully join with the joined tuples of s and t, we expect the final result to have nom copies.The order in which we perfor

11、m the natural join does not matter for bags.For sets:This is a similar case when dealing with bags except the joined tuples can only appear at most once in each result. If there are tuples r,s,t in relations R,S,T that can successfully join, then the result will contain a tuple with the schema of th

12、eir joined attributes.Exercise 5.1.4dFor bags:Suppose a tuple t occurs n and m times in bags R and S respectively. In the union of these two bags R S, tuple t would appear n + m times. Likewise, in the union of these two bags S R, tuple t would appear m + n times. Both sides of the relation yield th

13、e same result. For sets:A tuple t can only appear at most one time. Tuple t might appear each in sets R and S one or zero times. The combinations of number of occurrences for tuple t in R and S respectively are (0,0), (0,1), (1,0), and (1,1). Only when tuple t appears in both sets R and S will the u

14、nion R S have the tuple t. The same reasoning holds when we take the union S R. Therefore the commutative law for union holds.Exercise 5.1.4eFor bags:Suppose a tuple t occurs n and m times in bags R and S respectively. In the intersection of these two bags R S, tuple t would appear min( n,m ) times.

15、 Likewise in the intersection of these two bags S R, tuple t would appear min( m,n ) times. Both sides of the relation yield the same result. For sets:A tuple t can only appear at most one time. Tuple t might appear each in sets R and S one or zero times. The combinations of number of occurrences fo

16、r tuple t in R and S respectively are (0,0), (0,1), (1,0), and (1,1). Only when tuple t appears in at least one of the sets R and S will the intersection R S have the tuple t. The same reasoning holds when we take the intersection S R.Therefore the commutative law for intersection holds.Exercise 5.1

17、.4fFor bags:Suppose a tuple t occurs n times in bag R and tuple u occurs m times in bag S. Suppose also that the two tuples t,u can successfully join. Then in the natural join of these two bags R S, the joined tuple would appear nm times. Likewise in the natural join of these two bags S R, the joine

18、d tuple would appear mn times. Both sides of the relation yield the same result. For sets:An arbitrary tuple t can only appear at most one time in any set. Tuples u,v might appear respectively in sets R and S one or zero times. The combinations of number of occurrences for tuples u,v in R and S resp

19、ectively are (0,0), (0,1), (1,0), and (1,1). Only when tuple u exists in R and tuple v exists in S will the natural join R S have the joined tuple. The same reasoning holds when we take the natural join S R.Therefore the commutative law for natural join holds.Exercise 5.1.4gFor bags:Suppose tuple t

20、appears m times in R and n times in S. If we take the union of R and S first, we will get a relation where tuple t appears m + n times. Taking the projection of a list of attributes L will yield a resulting relation where the projected attributes from tuple t appear m + n times. If we take the proje

21、ction of the attributes in list L first, then the projected attributes from tuple t would appear m times from R and n times from S. The union of these resulting relations would have the projected attributes of tuple t appear m + n times.For sets:An arbitrary tuple t can only appear at most one time

22、in any set. Tuple t might appear in sets R and S one or zero times. The combinations of number of occurrences for tuple t in R and S respectively are (0,0), (0,1), (1,0), and (1,1). Only when tuple t exists in R or S (or both R and S) will the projected attributes of tuple t appear in the result.The

23、refore the law holds.Exercise 5.1.4hFor bags:Suppose tuple t appears u times in R, v times in S and w times in T. On the left hand side, the intersection of S and T would produce a result where tuple t would appear min(v , w) times. With the addition of the union of R, the overall result would have

24、u + min(v , w) copies of tuple t. On the right hand side, we would get a result of min(u + v, u + w) copies of tuple t. The expressions on both the left and right sides are equivalent.For sets:An arbitrary tuple t can only appear at most one time in any set. Tuple t might appear in sets R,S and T on

25、e or zero times. The combinations of number of occurrences for tuple t in R, S and T respectively are (0,0,0), (0,0,1), (0,1,0), (0,1,1), (1,0,0), (1,0,1), (1,1,0) and (1,1,1). Only when tuple t appears in R or in both S and T will the result have tuple t. Therefore the distributive law of union ove

26、r intersection holds.Exercise 5.1.4iSuppose that in relation R, u tuples satisfy condition C and v tuples satisfy condition D. Suppose also that w tuples satisfy both conditions C and D where w min(v , w). Then the left hand side will return those w tuples. On the right hand side, C(R) produces u tu

27、ples and D(R) produces v tuples. However, we know the intersection will produce the same w tuples in the result.When considering bags and sets, the only difference is bags allow duplicate tuples while sets only allow one copy of the tuple. The example above applies to both cases.Therefore the law ho

28、lds.Exercise 5.1.5aFor sets, an arbitrary tuple t appears on the left hand side if it appears in both R,S and not in T. The same is true for the right hand side.As an example for bags, suppose that tuple t appears one time each in both R,T and two times in S. The result of the left hand side would h

29、ave zero copies of tuple t while the right hand side would have one copy of tuple t. Therefore the law holds for sets but not for bags.Exercise 5.1.5bFor sets, an arbitrary tuple t appears on the left hand side if it appears in R and either S or T. This is equivalent to saying tuple t only appears w

30、hen it is in at least R and S or in R and T. The equivalence is exactly the right sides expression.As an example for bags, suppose that tuple t appears one time in R and two times each in S and T. Then the left hand side would have one copy of tuple t in the result while the right hand side would ha

31、ve two copies of tuple t.Therefore the law holds for sets but not for bags.Exercise 5.1.5cFor sets, an arbitrary tuple t appears on the left hand side if it satisfies condition C, condition D or both condition C and D. On the right hand side, C(R) selects those tuples that satisfy condition C while

32、D(R) selects those tuples that satisfy condition D. However, the union operator will eliminate duplicate tuples, namely those tuples that satisfy both condition C and D. Thus we are ensured that both sides are equivalent.As an example for bags, we only need to look at the union operator. If there ar

33、e indeed tuples that satisfy both conditions C and D, then the right hand side will contain duplicate copies of those tuples. The left hand side, however, will only have one copy for each tuple of the original set of tuples.Exercise 5.2.1aA+BA2B210154910164167916Exercise 5.2.1bB+1C-1103334431143Exer

34、cise 5.2.1cAB0101232434Exercise 5.2.1dBC010224253434Exercise 5.2.1eAB01232434Exercise 5.2.1fBC0124253402Exercise 5.2.1gASUM(B)022734Exercise 5.2.1hBAVG(C)01.524.534Exercise 5.2.1iA023Exercise 5.2.1jAMAX(C)24Exercise 5.2.1kABC23423401012434Exercise 5.2.1lABC23423401242502Exercise 5.2.1mABC23423401012

35、43401242502Exercise 5.2.1nAR.BS.BC012401250134013401240125013401342324340102Exercise 5.2.2aApplying the operator on a relation with no duplicates will yield the same relation. Thus is idempotent.Exercise 5.2.2bThe result of L is a relation over the list of attributes L. Performing the projection aga

36、in will return the same relation because the relation only contains the list of attributes L. Thus L is idempotent.Exercise 5.2.2cThe result of C is a relation where condition C is satisfied by every tuple. Performing the selection again will return the same relation because the relation only contai

37、ns tuples that satisfy the condition C. Thus C is idempotent.Exercise 5.2.2dThe result of L is a relation whose schema consists of the grouping attributes and the aggregated attributes. If we perform the same grouping operation, there is no guarantee that the expression would make sense. The groupin

38、g attributes will still appear in the new result. However, the aggregated attributes may or may not appear correctly. If the aggregated attribute is given a different name than the original attribute, then performing L would not make sense because it contains an aggregation for an attribute name tha

39、t does not exist. In this case, the resulting relation would, according to the definition, only contain the grouping attributes. Thus, L is not idempotent.Exercise 5.2.2eThe result of is a sorted list of tuples based on some attributes L. If L is not the entire schema of relation R, then there are a

40、ttributes that are not sorted on. If in relation R there are two tuples that agree in all attributes L and disagree in some of the remaining attributes not in L, then it is arbitrary as to which order these two tuples appear in the result. Thus, performing the operation multiple times can yield a di

41、fferent relation where these two tuples are s. Thus, is not idempotent.Exercise 5.2.3If we only consider sets, then it is possible. We can take A(R) and do a product with itself. From this product, we take the tuples where the two columns are equal to each other.If we consider bags as well, then it

42、is not possible. Take the case where we have the two tuples (1,0) and (1,0). We wish to produce a relation that contains tuples (1,1) and (1,1). If we use the classical operations of relational algebra, we can either get a result where there are no tuples or four copies of the tuple (1,1). It is not

43、 possible to get the desired relation because no operation can distinguish between the original tuples and the duplicated tuples. Thus it is not possible to get the relation with the two tuples (1,1) and (1,1).Exercise 5.3.1a) Answer(model) PC(model,speed,_,_,_) AND speed 3.00b) Answer(maker) Laptop

44、(model,_,_,hd,_,_) AND Product(maker,model,_) AND hd 100c) Answer(model,price) PC(model,_,_,_,price) AND Product(maker,model,_) AND maker=BAnswer(model,price) Laptop(model,_,_,_,_,price) AND Product(maker,model,_) AND maker=BAnswer(model,price) Printer(model,_,_,price) AND Product(maker,model,_) AND

45、 maker=Bd) Answer(model) Printer(model,color,type,_) AND color=true AND type=lasere) PCMaker(maker) Product(maker,_,type) AND type=pcLaptopMaker(maker) Product(maker,_,type) AND type=laptopAnswer(maker) LaptopMaker(maker) AND NOT PCMaker(maker)f) Answer(hd) PC(model1,_,_,hd,_) AND PC(model2,_,_,hd,_) AND model1 model2g) Answer(model1,model2) PC(model1,speed, ram,_,_) AND PC(model2,_speed,ram,_,_) AND model1 model2h) FastComputer(model) PC(model,speed,_,_,_) AND speed 2.80FastComputer(model) Laptop(model,speed,_,_,_,_) AND speed 2.80Answer(maker) Product(ma