数据库系统实现(第二版)答案ch6.docxVIP

Database System ImplementationSolutions for Chapter 6Solutions for Section 6.1Solutions for Section 6.3Solutions for Section 6.4Solutions for Section 6.5Solutions for Section 6.6Solutions for Section 6.7Solutions for Section 6.8Solutions for Section 6.9Solutions for Section 6.10Solutions for Section 6.1Exercise 6.1.1(a)Technically, the U_S operator assumes that the arguments are sets, as well as the result. Since the arguments are really bags, the best we can do is use the SQL union operator, which removes duplicates from the output even if there are duplicates in the input. The answer is {(0,1), (2,3), (2,4), (3,4), (2,5), (0,2)}.Exercise 6.1.1(i){(1,0,1), (5,4,9), (1,0,1), (6,4,16), (7,9,16)}.Exercise 6.1.1(k){(0,1), (0,1), (2,4), (3,4)}. Note that all five tuples of?R?satisfy the first condition, that the first component is less than the second. The last two tuples satisfy the condition that the sum of the components is at least 6, and the tuple (0,1) satisfies the condition that the sum exceed the product. Only the tuple (2,3) satisfies neither of the latter two conditions, and therefore doesnt appear in the result.Exercise 6.1.2(a){(0,1,2), (0,1,2), (0,1,2), (0,1,2), (2,3,5), (2,4,5), (3,4,5)}.Exercise 6.1.2(e){(0,1,4,5), (0,1,4,5), (2,3,2,3), (2,4,2,3)}.Exercise 6.1.2(g){(0,2), (2,7), (3,4)}.Exercise 6.1.3(a)One way to express the semijoin is by projecting?S?onto the list?L?of attributes in common with?R?and then taking the natural join:?R?JOIN pi_L(S).A second way is to take the natural join first, and then project onto the attributes of?R.Exercise 6.1.3(c)First, we need to find the tuples of?R?that dont join with?S?and therefore need to be padded. Let?T = R - pi_M(R JOIN S), where?M?is the schema, or list of attributes, of?R. Then the left outerjoin is?(R JOIN S) UNION (T x N), where?N?is the suitable ``null relation with one tuple that has?NULL?for each attribute of?Sthat is not an attribute of?R.Exercise 6.1.5(a)When a relation has no duplicate tuples, DEL