数据库实验三

实验要求：

1.以常用“名字大全”与“百家姓”数据集为基础，生成不小于1千万条stud记录，要求，姓名的重复率不超过10%，学号以ABCD16EFGH为格式模板，即其中16是固定的，AB为从01到80，CD为从01到90，EF为01到50，GH为01到32；性别中，男、女占比为99%到99.5%。TEL与E-mail不作要求，但不能全空。Birthday要求从‘19940101’到‘19990731’分布。要求记录ORACLE数据文件的大小变化。（需要编制过程）

imp U_J122/U_J123 file = E:大三上大型数据库大型数据库
ame.dmp fromuser = yjs touser = U_J122 commit=y

drop table name1;
create table name1(word1 NCHAR(1));
drop table name2;
create table name2(word2 NCHAR(1));
drop table name3;
create table name3(word3 NCHAR(1));
insert into name1 select distinct substr(name,1,1) from name;
insert into name2 select distinct substr(name,2,1) from name;
insert into name3 select distinct substr(name,3,1) from name WHERE ROWNUM <=12;

create table T_namex_j122(namex VARCHAR2(10));
insert into T_namex_j122 (select word1||word2||word3 nameitem from name_1st,name_2nd,name_3rd);

drop table T_studnew_j122;
create table T_studnew_j122(
    sno CHAR(10) primary key,
    sname VARCHAR2(10),
    sex VARCHAR2(4),
    tel VARCHAR2(15),
    email VARCHAR2(30),
    birthday DATE
);
set TIMING on;
drop procedure P_create_stud_j122;
create OR REPLACE PROCEDURE P_create_stud_j122 IS
    cnt PLS_INTEGER:=1;
    msno CHAR(10);
    mname VARCHAR2(10);
    msex VARCHAR2(4);
    mtel VARCHAR2(15);
    memail VARCHAR2(30);
    mbirthday DATE;

    maxl PLS_INTEGER:=10322736;
    yy PLS_INTEGER;
    mm PLS_INTEGER;
    dd PLS_INTEGER;

    rand_sex PLS_INTEGER;
    rand_eml varchar2(30);

    CURSOR c_name IS (select namex from T_namex_j122);
BEGIN
    
    OPEN c_name;
    FOR ab IN 1..80 
        LOOP EXIT WHEN cnt>maxl;
        FOR cd IN 1..90
            LOOP EXIT WHEN cnt>maxl;
            FOR ef IN 1..50 
                LOOP EXIT WHEN cnt>maxl;
                FOR gh IN 1..32 
                    LOOP EXIT WHEN cnt>maxl;

        msno:= TO_CHAR(ab,'FM09')||TO_CHAR(cd,'FM09')||'16'||TO_CHAR(ef,'FM09')||TO_CHAR(gh,'FM09');

        FETCH c_name INTO mname;

        rand_sex:=DBMS_RANDOM.VALUE(1,100);
        IF rand_sex<=40 THEN msex:='男';
            ELSIF rand_sex<=99 THEN msex:='女';
            ELSE msex:='其他';
        END IF;

        rand_eml:=DBMS_RANDOM.STRING('X',10);
        memail:=rand_eml||'@163.com';

        mtel:=ab||cd||ef||gh;

        yy:=dbms_random.value(1994,1999);
        IF yy<1999 THEN 
            mm:=dbms_random.value(1,12);
            ELSE mm:=dbms_random.value(1,7);
        END IF;
        IF mm=2 AND ((0=yy mod 4 AND 0!=yy mod 100) OR 0=yy mod 400)THEN
            dd:=dbms_random.value(1,29);
            ELSIF mm=2 THEN
                dd:=dbms_random.value(1,28);
            ELSIF mm=1 OR mm=3 OR mm=5 OR mm=7 OR mm=8 OR mm=10 OR mm=12 THEN
                dd:=dbms_random.value(1,31);
            ELSE dd:=dbms_random.value(1,30);
        END IF;
        mbirthday:=TO_DATE(yy||TO_CHAR(mm,'FM09')||TO_CHAR(dd,'FM09'),'yyyymmdd');

        INSERT INTO T_studnew_j122(sno,sname,sex,email,tel,birthday)
        VALUES (msno,mname,msex,memail,mtel,mbirthday);

        cnt:=cnt+1;
                END LOOP;
            END LOOP;
        END LOOP;
    END LOOP;
CLOSE c_name;
END;
.
/

exec P_create_stud_j122;

2．分别测试stud有主键与没有主键情形下生成记录的时间。

3．建立基于name的索引index_name，测试建立的时间与建立索引前后查询某一姓名及某一姓的时间长度。

4．测试索引index_name建立前后，分姓（简单地理解为姓名的第1，2位）的记录数统计时间。

select * from T_studnew_j122 where sno='0101160101';
select * from T_studnew_j122 where sname='任盈强';
select * from T_studnew_j122 where sname like '任%';

create index index_name on T_studnew_j122(sname);
select * from T_studnew_j122 where sname='任盈强';
select * from T_studnew_j122 where sname like '任%';

5．按学号首位建立10个分区分别为part_0到part_9，测试建立分区前后分首位统计人数与分专业（EF位）统计人数的时间差别。

DROP TABLE T_studnew_j122;
CREATE TABLE T_studnew_j122(
    sno CHAR(10),
    sname VARCHAR2(10),
    sex VARCHAR2(4),
    tel VARCHAR2(15),
    email VARCHAR2(30),
    birthday DATE
)partition by range(sno)(
partition part_0 values less than ('1000000000'),
partition part_1 values less than ('2000000000'),
partition part_2 values less than ('3000000000'),
partition part_3 values less than ('4000000000'),
partition part_4 values less than ('5000000000'),
partition part_5 values less than ('6000000000'),
partition part_6 values less than ('7000000000'),
partition part_7 values less than ('8000000000')
);

select count(*) from T_studnew_j122 partition(part_0);
select count(*) from T_studnew_j122 partition(part_1);
select count(*) from T_studnew_j122 partition(part_2);
select count(*) from T_studnew_j122 partition(part_3);
select count(*) from T_studnew_j122 partition(part_4);
select count(*) from T_studnew_j122 partition(part_5);
select count(*) from T_studnew_j122 partition(part_6);
select count(*) from T_studnew_j122 partition(part_7);