下面看看对其他Oracle特有的语法支持的如何：
SQL>SELECT *
2 FROM T, T1
3 WHERE T.ID = T1.ID;
SELECT *
FROM T, T1
WHERE T.ID = T1.ID;
id name create_date ID
1 1 TEST 2010-03-31 1
1 rows got
time used: 0.521(ms) clock tick:855860.
SQL>SELECT *
2 FROM T, T1
3 WHERE T.ID = T1.ID (+);
SELECT *
FROM T, T1
WHERE T.ID = T1.ID (+);
id name create_date ID
1 1 TEST 2010-03-31 1
2 2 NEWNAME 2010-03-31 NULL
3 2 TEST 2010-03-31 NULL
3 rows got
time used: 36.644(ms) clock tick:61195700.
2 FROM T, T1
3 WHERE T.ID = T1.ID;
SELECT *
FROM T, T1
WHERE T.ID = T1.ID;
id name create_date ID
1 1 TEST 2010-03-31 1
1 rows got
time used: 0.521(ms) clock tick:855860.
SQL>SELECT *
2 FROM T, T1
3 WHERE T.ID = T1.ID (+);
SELECT *
FROM T, T1
WHERE T.ID = T1.ID (+);
id name create_date ID
1 1 TEST 2010-03-31 1
2 2 NEWNAME 2010-03-31 NULL
3 2 TEST 2010-03-31 NULL
3 rows got
time used: 36.644(ms) clock tick:61195700.
不但普通的连接支持，连Oracle的外连接’+’都是支持的。
SQL>SELECT *
2 FROM
3 (
4 SELECT ROWNUM RN, A.*
5 FROM
6 (
7 SELECT T.*
8 FROM T, T1
9 WHERE T.ID = T1.ID(+)
10 ORDER BY 2, 1
11 ) A
12 WHERE ROWNUM <= 5
13 )
14 WHERE RN > 1;
SELECT *
FROM
(
SELECT ROWNUM RN, A.*
FROM
(
SELECT T.*
FROM T, T1
WHERE T.ID = T1.ID(+)
ORDER BY 2, 1
) A
WHERE ROWNUM <= 5
)
WHERE RN > 1;
RN id name create_date
1 2 1 TEST 2010-03-31
2 3 2 TEST 2010-03-31
2 rows got
time used: 30.910(ms) clock tick:51615050.
2 FROM
3 (
4 SELECT ROWNUM RN, A.*
5 FROM
6 (
7 SELECT T.*
8 FROM T, T1
9 WHERE T.ID = T1.ID(+)
10 ORDER BY 2, 1
11 ) A
12 WHERE ROWNUM <= 5
13 )
14 WHERE RN > 1;
SELECT *
FROM
(
SELECT ROWNUM RN, A.*
FROM
(
SELECT T.*
FROM T, T1
WHERE T.ID = T1.ID(+)
ORDER BY 2, 1
) A
WHERE ROWNUM <= 5
)
WHERE RN > 1;
RN id name create_date
1 2 1 TEST 2010-03-31
2 3 2 TEST 2010-03-31
2 rows got
time used: 30.910(ms) clock tick:51615050.
Oracle的标准分页语句是支持的，查询子查询语句是支持的，ORDER BY常量代替列名也是支持的。
SQL>SELECT ROW_NUMBER() OVER(ORDER BY NAME, ID DESC) RN,
2 ID,
3 NAME
4 FROM T;
第1行: '('附近有语法错误
2 ID,
3 NAME
4 FROM T;
第1行: '('附近有语法错误
显然分析函数是不支持的，虽然分析函数在8i就出现了，不过确实属于Oracle比较特有的技术，达梦不支持分析函数并不意外。
SQL>SELECT NAME, MAX(ID)
2 FROM T
3 GROUP BY NAME
4 HAVING COUNT(*) > 1;
SELECT NAME, MAX(ID)
FROM T
GROUP BY NAME
HAVING COUNT(*) > 1;
NAME
1 TEST 2
1 rows got
time used: 0.583(ms) clock tick:963240.
SQL>SELECT NAME, AVG(ID)
2 FROM T
3 GROUP BY ROLLUP(NAME);
SELECT NAME, AVG(ID)
FROM T
GROUP BY ROLLUP(NAME);
无效的存储过程名 'ROLLUP' .error code = -1024
2 FROM T
3 GROUP BY NAME
4 HAVING COUNT(*) > 1;
SELECT NAME, MAX(ID)
FROM T
GROUP BY NAME
HAVING COUNT(*) > 1;
NAME
1 TEST 2
1 rows got
time used: 0.583(ms) clock tick:963240.
SQL>SELECT NAME, AVG(ID)
2 FROM T
3 GROUP BY ROLLUP(NAME);
SELECT NAME, AVG(ID)
FROM T
GROUP BY ROLLUP(NAME);
无效的存储过程名 'ROLLUP' .error code = -1024
显然聚集函数、GROUP BY语句和HAVING语句都是支持的。但是8i开始Oracle对GROUP BY提供了ROLLUP和CUBE功能，而这显然是达梦所不支持的，既然ROLLUP和CUBE不支持，那么对应的GROUPING等CUBE和ROLLUP专用的函数肯定也是不支持的。
SQL>SELECT T.ID, T1.ID ID1
2 FROM T, T1
3 WHERE T.ID = T1.ID (+)
4 ORDER BY 2 NULLS FIRST;
第4行: 'NULLS'附近有语法错误
2 FROM T, T1
3 WHERE T.ID = T1.ID (+)
4 ORDER BY 2 NULLS FIRST;
第4行: 'NULLS'附近有语法错误
支持ORDER BY语句，但是Oracle的NULLS LAST和NULLS FIRST语法并不支持。
SQL>SELECT ID, (SELECT ID FROM T1 WHERE T1.ID = T.ID)
2 FROM T;
SELECT ID, (SELECT ID FROM T1 WHERE T1.ID = T.ID)
FROM T;
ID
1 1 1
2 2 NULL
3 2 NULL
3 rows got
time used: 0.556(ms) clock tick:916730.
2 FROM T;
SELECT ID, (SELECT ID FROM T1 WHERE T1.ID = T.ID)
FROM T;
ID
1 1 1
2 2 NULL
3 2 NULL
3 rows got
time used: 0.556(ms) clock tick:916730.
