Topic 8 Structured Query Language (SQL) : DML Part 2

Size: px

Start display at page:

Download "Topic 8 Structured Query Language (SQL) : DML Part 2"

Esmond Anthony
5 years ago
Views:

1 FIT1004 Database Topic 8 Structured Query Language (SQL) : DML Part 2 Learning Objectives: Use SQL functions Manipulate sets of data Write subqueries Manipulate data in the database References: Rob, P. & Coronel, C., Database Systems, 6 th Edition, Chapters. 6 & 7 Rob, P. & Coronel, C., Database Systems, 7 th Edition, Chapters. 7 &

2 Where We Are Introduction to Database Systems The Relational Model Database Lifecycle Conceptual Design Logical Design Normalisation Physical Design Implementation SQL (DML) SQL (DDL & DCL) Transaction Management Database Administration Data Warehousing & Data Mining 2

3 SQL Functions are useful for manipulating data by decomposing the data elements use numerical, date or string values and may appear in a SQL statement wherever a value or attribute is used support within a DBMS differs between vendors are categorised according to their operand types Function Type Applicable To Arithmetic numerical data Text alphanumeric date Date date/time-related data General any datatype Conversion datatype conversion Group sets of values 3

4 Oracle Arithmetic Functions abs(n) The column s absolute value ceil(n) Nearest whole integer greater than or equal to number > SELECT ceil(10.6) FROM dual; floor(n) Largest integer equal to or less than n > SELECT floor(10.6) FROM dual; mod(m,n) Remainder of m divided by n. If n=0, then m is returned > SELECT mod(7,5) FROM dual; power(m,n) Number m raised to the power of n round(n,m) Results rounded to m places to the right of decimal point sign(n) If n=0, returns 0; if n>0, returns 1; if n<0, returns -1 sqrt(n) Square root of n 4

5 Oracle Text Functions initcap(char) Changes the first character of each character string to uppercase > SELECT initcap( mr teplow ) FROM dual; lower(char), upper(char) Makes the entire string lowercase/uppercase > SELECT lower(ename) FROM employee; replace(char, str1, str2) Character string with every occurrence of str1 being replaced with str2 substr(char,m,n) Picks off part of the character string char starting in position m for n characters > SELECT substr( ABCDEF,2,1) FROM dual; length(char) Length of char > SELECT length( Anderson ) FROM dual; str1 str2 Concatenates two character fields together ltrim(char[, k]), rtrim(char[, k]) remove characters from the left/right of t, until the first character not in k 5

6 Oracle Date Functions last_day Last day of the month SELECT last_day(sysdate) FROM dual; add_months(d,n) Adds or subtracts n months from date d SELECT add_months(sysdate, 2) FROM dual; months_between(f,s) Difference in months between date f and date s SELECT months_between(sysdate, '1-JAN-2006') FROM dual; next_day(d,day) Date that is the specified day of the week after d SELECT next_day(sysdate, 'Monday') FROM dual; extract(c from d) Extract date/time component c from expression d SELECT bdate, extract (year from bdate) AS year_of_birth, extract (month from bdate) AS month_of_birth, extract (day from bdate) AS day_of_birth FROM employee; 6

7 Oracle General Functions greatest(a, b, ) greatest value of the function arguments least(a, b, ) least value of the function arguments SELECT greatest(12*6, 148/2, 73), least(12*6, 148/2, 73) FROM dual; nullif(a, b) NULL if a = b; otherwise a NVL(x, y) y if x is NULL; otherwise x decode (x, a 1, b 1, a 2, b 2,., a n, b n [, y]) b 1 if x = a 1, b 2 if x = a 2,. b n if x = a n, and otherwise y (or default:null) 7

8 Oracle Conversion Functions Three main conversion functions: to_char > converts any data type to character data to_number > converts a valid set of numeric character data to number data type SELECT 123, to_char(123, '$ '), to_number('123') FROM dual; to_date > converts character data of the proper format to date data type. 8

9 Conversion Functions Yor YY or YYYY Last one, two,three or four digits of year. Q Quarter of year (Jan thru March = 1) MM, RM Month(01-12), Roman numeral month (IV for April) Month Name of month WW, W Week of year, Week of month DDD, DD, D Day of the year, month, week DY Abbreviated name of day HH or HH12 Hour of day HH24 Hour of day using 24-hour clock MI Minutes (0-59) SS Seconds (0-59) 9

10 Conversion Functions Yor YY or YYYY Last one, two,three or four digits of year. > SELECT to_char(sysdate, YYYY ) FROM dual; Q Quarter of year (Jan thru March = 1) > SELECT to_char(sysdate, Q ) FROM dual; MM, RM Month(01-12), Roman numeral month (IV for April) > SELECT to_char(sysdate, MM) FROM dual; > SELECT to_char(sysdate, RM ) FROM dual; Month Name of month > SELECT to_char(sysdate, Month ) FROM dual; WW, W Week of year, Week of month > SELECT to_char(sysdate, WW ) FROM dual; > SELECT to_char(sysdate, W ) FROM dual; 10

11 Conversion Functions DDD, DD, D Day of the year, month, week > SELECT to_char(sysdate, DDD ) FROM dual; > SELECT to_char(sysdate, DD ) FROM dual; > SELECT to_char(sysdate, D ) FROM dual; DY, DAY Abbreviated, full name of day > SELECT to_char(sysdate, DY ) FROM dual; HH or HH12 Hour of day > SELECT to_char(sysdate, HH ) FROM dual; HH24 Hour of day using 24-hour clock MI Minutes (0-59) SS Seconds (0-59) > SELECT to_char(sysdate, HH24:MI:SS ) FROM dual; 11

12 Handling of dates in Oracle Dates are stored differently from the SQL standard standard uses two different types: date and time Oracle uses one type: DATE > Stored in internal format contains date and time > Output is controlled by formatting select to_char(sysdate,'dd-mon-yyyy') from dual;» 25-Aug-2006 select to_char(sysdate,'dd-mon-yyyy hh:mi:ss pm') from dual;» 25-Aug :56:50 AM 10G introduced TIMESTAMP finer granularity on secs > select cast(sysdate as TIMESTAMP) from dual; 25/AUG/06 11:15: AM 12

13 Handling of dates in Oracle cont d DATE data type should normally be formatted with TO_CHAR when selecting for display formatted with TO_DATE when comparing or inserting/updating Example: select empno, ename, to_char(bdate,'dd-mon-yyyy') from payroll.employee where bdate > to_date('01-feb-1962','dd-mon-yyyy') order by bdate In the where clause, why not convert bdate on the left to CHAR? where to_char(bdate,'dd-mon-yyyy') > '01-Feb-1962' 13

14 Oracle Group Functions Used to perform mathematical summaries such as: counting the number of rows finding the minimum and maximum values for some specified attribute summing the values in a column, and averaging the values in a specified column. All group functions can be applied only to sets of values return a single aggregated value, derived from a set of values NULL values are ignored by the group functions, the only exception is the COUNT(*) function 14

15 COUNT Function Can use the COUNT function for any datatype COUNT(*) returns the number of rows in a table SELECT COUNT(*) FROM employee; COUNT(expr) returns the number of nonnull rows. SELECT COUNT(empno) FROM employee; SELECT COUNT(comm) FROM employee; 15

16 MAX and MIN Functions Use MIN and MAX functions for any datatype. SELECT MIN(bdate), MAX(bdate) FROM employee; SELECT MIN(ename), MAX(ename) FROM employee; SELECT MIN(comm), MAX(comm) FROM employee; 16

17 SUM and AVG Functions Use the SUM and AVG functions for numeric data. SELECT SUM(msal) FROM employee; SELECT SUM(comm) FROM employee; SELECT AVG(msal) FROM employee; SELECT AVG(comm) FROM employee; 17

18 NVL Function The NVL function forces group functions to include null values. SELECT COUNT(NVL(comm, 0)) FROM employee; SELECT AVG(NVL(comm, 0)) FROM employee; NVL(x,, y) > y if x is NULL; otherwise x > y can be any value SELECT AVG(NVL(comm, 10)) FROM employee; 18

19 GROUP BY clause Is used to divide the rows in a table into groups Allows the aggregate functions to return summary information for each group All columns in the SELECT list that are not in the group functions must be included in the GROUP BY clause The GROUP BY column does not have to be in the SELECT clause SELECT deptno, COUNT(*), MIN(msal), MAX(msal), SUM(msal), AVG(msal) FROM employee GROUP BY deptno; 19

20 GROUP BY clause All columns in the SELECT list that are not in group functions must be in the GROUP BY clause. SELECT deptno, AVG(msal) FROM employee; SQL> SELECT deptno, AVG(msal) 2 FROM employee; SELECT deptno, AVG(msal) * ERROR at line 1: ORA-00937: not a single-group group function SELECT d.deptno, d.dname, AVG(e.msal) FROM employee e, department d WHERE e.deptno = d.deptno GROUP BY d.deptno; SQL> SELECT d.deptno, d.dname, AVG(e.msal) 2 FROM employee e, department d 3 WHERE e.deptno = d.deptno 4 GROUP BY d.deptno; SELECT d.deptno, d.dname, AVG(e.msal) * ERROR at line 1: ORA-00979: not a GROUP BY expression 20

21 GROUP BY clause The GROUP BY column does not have to be in the SELECT list. SELECT d.dname, AVG(e.msal) FROM employee e, department d WHERE e.deptno = d.deptno GROUP BY d.deptno, d.dname; Grouping by more than one column SELECT deptno, job, SUM(msal) FROM employee GROUP BY deptno, job; 21

22 Illegal queries using GROUP BY Who earns more than the average salary? SELECT empno, msal FROM employee WHERE msal > avg(msal); SQL> SELECT empno, msal 2 FROM employee 3 WHERE msal > avg(msal); WHERE msal > avg(msal) * ERROR at line 3: ORA-00934: group function is not allowed here Which department pays more than $10000 in salary per month? SELECT deptno, SUM(msal) FROM employee WHERE SUM(msal) > GROUP BY deptno; Cannot use the WHERE clause to restrict groups. 22

23 HAVING clause Is applied to the output of a GROUP BY operation to restrict the selected rows. Operates like a WHERE clause, however, the WHERE clause applies to columns and expressions for individual rows, while the HAVING clause is applied to the output of a GROUP BY operation. The DBMS evaluates the clauses in a SQL statement in the following order: WHERE clause GROUP BY clause HAVING clause Therefore if we wish to restrict the results of a query based on the result of a GROUP BY clause we need to use a HAVING clause rather than the WHERE clause. 23

24 HAVING clause Use the HAVING clause to restrict groups Rows are grouped. The group function is applied. Groups matching the HAVING clause are displayed. SQL> SELECT deptno, max(msal) 2 FROM employee 3 WHERE max(msal)> GROUP BY deptno; SQL> SELECT deptno, max(msal) 2 FROM employee 3 GROUP BY deptno 4 HAVING max(msal)>2900; WHERE max(msal)>2900 * ERROR at line 3: ORA-00934: group function is not allowed here DEPTNO MAX(MSAL)

25 HAVING clause SELECT job, SUM(msal) PAYROLL FROM employee WHERE job NOT LIKE 'SALES%' GROUP BY job HAVING SUM(msal)>5000 ORDER BY SUM(msal); SQL> SELECT job, SUM(msal) PAYROLL 2 FROM employee 3 WHERE job NOT LIKE 'SALES%' 4 GROUP BY job 5 HAVING SUM(msal)> ORDER BY SUM(msal); JOB PAYROLL TRAINER 7900 MANAGER

26 Nesting Functions Can nest aggregate functions Display the maximum average monthly salary SELECT max(avg(msal)) FROM employee GROUP BY deptno; Which department has the greatest average monthly salary? 26

27 Subqueries A subquery is a query that is embedded (or nested) inside another query Also known as a nested query or an inner query. SELECT select_list FROM table WHERE expr operator (SELECT select_list FROM table); The first query in the SQL statement is known as the outer query. The query inside the SQL statement is known as the inner query. The inner query is evaluated first and the output from this query is used as the input for the outer query. The inner query is normally expressed inside parentheses. 27

28 Subqueries The Oracle Server executes subqueries first. returns results into the clause of the main query. Which employee has a greater monthly salary than Blake? SELECT empno, ename, msal FROM employee WHERE msal > (SELECT msal FROM employee WHERE ename = 'BLAKE'); Guidelines Enclose subqueries in parentheses. Place subqueries on the right side of the comparison operator. Do not use an ORDER BY clause to a subquery. Use single-row operators with single-row subqueries. Use multiple-row operators with multiple-row subqueries 28

29 Types of Subqueries Single-row subquery Main query Subquery returns CLERK Multiple-row subquery Main query Subquery returns Multiple-column subquery Main query Subquery returns CLERK MANAGER CLERK 7900 MANAGER

30 Single-Row Subqueries Return only one row Use single-row comparison operators Operator = > >= < <= <> Meaning Equal to Greater than Greater than or equal to Less than Less than or equal to Not equal to 30

31 Single-Row Subqueries Display the name and job of employees who have the same job as Allen and a monthly salary greater than employee Ward SELECT ename, job, msal FROM employee WHERE job = (SELECT job FROM employee WHERE ename = 'ALLEN') AND msal > (SELECT msal FROM employee WHERE ename = 'WARD'); Display the name, job and monthly salary of employees who earn the least in salary. SELECT ename, job, msal FROM employee WHERE msal = (SELECT min(msal) FROM employee); 31

32 Single-Row Subqueries Which department has the most employees? SELECT d.deptno, d.dname, count(*) FROM department d, employee e WHERE d.deptno = e.deptno HAVING count(*) = (SELECT max(count(*)) FROM employee GROUP BY deptno) GROUP BY d.deptno, d.dname; Which department has the greatest average monthly salary? SELECT d.deptno, d.dname, avg(msal) FROM department d, employee e WHERE d.deptno = e.deptno HAVING avg(msal) = (SELECT max(avg(msal)) FROM employee GROUP BY deptno) GROUP BY d.deptno, d.dname; 32

33 Single-Row Subqueries What is wrong with this statement? SELECT empno, ename FROM employee WHERE msal = (SELECT MIN(msal) FROM employee GROUP BY deptno); SQL> SELECT empno, ename 2 FROM employee 3 WHERE msal = (SELECT MIN(msal) 4 FROM employee 5 GROUP BY deptno); WHERE msal = (SELECT MIN(msal) * ERROR at line 3: ORA-01427: single-row subquery returns more than one row 33

34 Multiple-Row Subqueries Return more than one row Use multiple-row comparison operators Operator IN ANY ALL Meaning Equal to any member in the list Compare value to each value returned by the subquery Compare value to every value returned by the subquery 34

35 Multiple-Row Subqueries Which employees will be displayed? SELECT empno, ename, job, msal FROM employee WHERE msal < ANY (SELECT msal FROM employee WHERE job = 'SALESREP') AND job <> 'SALESREP'; SQL> SELECT msal FROM employee 2 WHERE job = 'SALESREP'; MSAL

36 Multiple-Row Subqueries Which employees will be displayed? SELECT empno, ename, job, msal FROM employee WHERE msal > ALL (SELECT AVG(msal) FROM employee GROUP BY deptno); SQL> SELECT AVG(msal) FROM employee 2 GROUP BY deptno; AVG(MSAL)

37 Multiple-Column Subqueries The number of columns in the main query must match the number of columns returned from the inner query. Display the employees that work in the same department and have the same job as Martin. SELECT empno, ename, deptno, job FROM employee WHERE (deptno, job) = (SELECT deptno, job FROM employee WHERE ename = 'MARTIN'); Write another SQL query that would produce the same result. 37

38 Relational Set Operators SQL data manipulation commands are set oriented, that is they operate over entire sets of rows and columns at once. Using the set operators you can combine two or more sets to create new sets (relations) Union All rows selected by either query Union All All rows selected by either query, including all duplicates Intersect All distinct rows selected by both queries Minus All distinct rows selected by the first query but not the second All set operators have equal precedence. If a SQL statement contains multiple set operators, Oracle evaluates them from the left to right if no parentheses explicitly specify another order. The corresponding expressions in the select lists of the component queries of a compound query must match in number and datatype. 38

39 UNION The UNION statement combines rows from two or more queries without including duplicate rows. The UNION ALL statement combines rows from two or more queries and retains the duplicate rows. The following statement combines the results with the UNION operator, which eliminates duplicate selected rows. You must match datatype (using the TO_CHAR, TO_DATE and TO_NUMBER functions) when columns do not exist in one or the other table: SELECT 'Manager', empno, ename, job, mgr FROM employee WHERE empno IN (SELECT mgr FROM employee) UNION SELECT 'Employee', empno, ename, job, mgr FROM employee WHERE empno NOT IN (SELECT NVL(mgr, 0) FROM employee) ORDER BY mgr 39

40 INTERSECT The INTERSECT statement combines rows from two queries and returns only those rows that appear in both sets. SELECT empno, ename, job, mgr FROM employee WHERE empno IN (SELECT mgr FROM employee) INTERSECT SELECT empno, ename, job, mgr FROM employee; EMPNO ENAME JOB MGR JONES MANAGER BLAKE MANAGER CLARK MANAGER SCOTT TRAINER KING DIRECTOR 7902 FORD TRAINER rows selected. 40

41 MINUS The MINUS statement combines rows from two queries and returns only those rows that appear in the first set but not in the second. SELECT empno, ename, job, mgr FROM employee MINUS SELECT empno, ename, job, mgr FROM employee WHERE empno IN (SELECT mgr FROM employee); EMPNO ENAME JOB MGR SMITH TRAINER ALLEN SALESREP WARD SALESREP MARTIN SALESREP TURNER SALESREP ADAMS TRAINER JONES ADMIN MILLER ADMIN rows selected. 41

42 Manipulating data There are six basic SQL data manipulation commands 42

43 INSERT statement The INSERT statement is used to enter data into a table INSERT INTO table [(column [, column...])] VALUES (value [, value...]); The INSERT statement allows the insertion of data one row at a time If you insert a new row that contains values for each column in the table, the column list is not required in the INSERT clause If you do not use the column list, the values must be listed according to the default order of the columns in the table INSERT INTO department VALUES (50, 'SUPPORT', 'SEATTLE', 7788); INSERT INTO VALUES department (deptno, dname, location) (60, 'ADMIN', 'ORLANDO'); 43

44 INSERT statement Can use the reserved word NULL to specify a null value for a specific column INSERT INTO department VALUES (50, 'SUPPORT', 'SEATTLE', NULL); Can specify the reserved word DEFAULT to insert the default value associated with the corresponding column. If a DEFAULT value was not defined for the column a NULL is inserted instead INSERT INTO employee (empno, ename, init, bdate, msal, deptno) VALUES (7999, 'DUCK', 'D.', to_date('01-jun-1985, DD-MON-YYYY ), 0, DEFAULT); What is the problem with this statement? INSERT INTO employee (empno, ename, init, bdate, msal, deptno) VALUES (8999, 'DOO', 'S.', to_date('01-jun-1985, DD-MON-YYYY ), 4995, 60); What is the problem with this statement? INSERT INTO employee (empno, ename, init, bdate, msal, deptno) VALUES (8999, 'O'BRIEN', 'F.', to_date('01-jun-1985, DD-MON-YYYY ), 4995, 60); 44

45 INSERT statement Can use subqueries in the VALUES clause INSERT INTO employee(empno, ename, init, bdate, msal) VALUES (1111, 'Baggins', 'B.', to_date('25-nov-1985, DD-MON-YYYY ), (SELECT msal FROM employee WHERE ename = 'SMITH')); To insert multiple rows of data use an INSERT statement with a subquery. INSERT INTO table [ column (, column) ] (subquery); INSERT INTO managers(id, name, salary, bdate) SELECT empno, ename, msal, bdate FROM employee WHERE job = 'MANAGER'; Do not use the VALUES clause. Match the number of columns in the INSERT clause to those in the subquery. 45

46 UPDATE statement The UPDATE statement allows you to change attribute values in one or more rows of a table. UPDATE table SET column = value [, column = value,...] [WHERE condition]; UPDATE: the table you want to update SET: the change you want to apply WHERE: the rows to which you want to apply the change if you omit the optional WHERE clause the change is applied to all rows of the table 46

47 UPDATE statement UPDATE employee SET job = 'SALESREP', msal = msal - 500, comm = 0, deptno = 30 WHERE empno = 7876; What is the problem with this statement? UPDATE employee SET deptno = 50 WHERE empno = 7876; 47

48 UPDATE statement You can use subqueries in an UPDATE statement to update rows in a table based on values from another table. UPDATE table SET column = (subquery) [, column = value,...] [WHERE condition]; UPDATE employee SET job = (SELECT job FROM employee WHERE ename = 'CLARK'), deptno = (SELECT deptno FROM employee WHERE ename = 'ALLEN') WHERE empno = 7876; 48

49 DELETE statement The DELETE statement allows you to delete rows of data from a table. DELETE FROM table [WHERE condition]; the WHERE clause is optional, if omitted the DELETE command will delete all rows in the table DELETE FROM employee; DELETE FROM employee WHERE empno = 7999; What is the problem with this statement? DELETE FROM employee WHERE empno = 7566; 49

50 DELETE statement You can use subqueries in a DELETE statement to delete rows in a table based on values from another table. DELETE FROM table [WHERE condition = (subquery)]; DELETE FROM employee WHERE deptno = (SELECT deptno FROM department WHERE dname = 'SALES'); 50

51 Summary This lecture SQL operators and functions Manipulating sets of data Writing subqueries Manipulating data in the database Next lecture Understand the datatypes supported by the SQL standard Use DDL statements to create, alter and drop database objects Use DCL statements to control access to database objects Understand the importance of transaction management Understand the importance of concurrency control Understand the importance of database recovery 51

CS2 Current Technologies Lecture 2: SQL Programming Basics

T E H U N I V E R S I T Y O H F R G E D I N B U CS2 Current Technologies Lecture 2: SQL Programming Basics Dr Chris Walton (cdw@dcs.ed.ac.uk) 4 February 2002 The SQL Language 1 Structured Query Language