USER_TAB_COLS和USER_TAB_COLUMNS两个视图有什么区别.

看oracle的元数据，

--by yl landgis@126.com,yanleigis@21cn.com 2008.9.12

create or replace view sys.user_tab_columns as
select TABLE_NAME, COLUMN_NAME, DATA_TYPE, DATA_TYPE_MOD, DATA_TYPE_OWNER,
       DATA_LENGTH, DATA_PRECISION, DATA_SCALE, NULLABLE, COLUMN_ID,
       DEFAULT_LENGTH, DATA_DEFAULT, NUM_DISTINCT, LOW_VALUE, HIGH_VALUE,
       DENSITY, NUM_NULLS, NUM_BUCKETS, LAST_ANALYZED, SAMPLE_SIZE,
       CHARACTER_SET_NAME, CHAR_COL_DECL_LENGTH,
       GLOBAL_STATS, USER_STATS, AVG_COL_LEN, CHAR_LENGTH, CHAR_USED,
       V80_FMT_IMAGE, DATA_UPGRADED, HISTOGRAM
  from USER_TAB_COLS
 where HIDDEN_COLUMN = 'NO'

HIDDEN_COLUMN 字段来源

SQL> CREATE TABLE yl (id integer,name varchar2(32));

Table created

SQL> INSERT INTO yl SELECT  rownum,'yanleigis' || LPAD(ROWNUM,2,'0') FROM DUAL CONNECT BY ROWNUM<=11;

11 rows inserted

SQL> commit;

Commit complete

SQL> CREATE INDEX idx_Name_yl ON yl(SUBSTR(name,4));

Index created

SQL> DESC yl;
Name Type                 Nullable Default Comments
-------- -------------         -----------  ---------- ---------------
ID        INTEGER               Y                        
NAME VARCHAR2(32)    Y                        

SQL> select * from yl;

     ID                  NAME
---------------- ----------------------
     1                  yanleigis01
     2                  yanleigis02
     3                  yanleigis03
     4                  yanleigis04
     5                  yanleigis05
     6                  yanleigis06
     7                  yanleigis07
     8                  yanleigis08
     9                  yanleigis09
     10                yanleigis10
     11                yanleigis11

11 rows selected

SQL> SELECT COLUMN_NAME,HIDDEN_COLUMN FROM user_TAB_COLS WHERE TABLE_NAME='YL';

COLUMN_NAME           HIDDEN_COLUMN
-------------------------        ---------------------------
ID                                      NO
NAME                               NO
SYS_NC00003$            YES

SQL>  SELECT SYS_NC00003$ ,id,name FROM yl;

SYS_NC00003$       ID                NAME
--------------------------  -------------  ---------------------
leigis01                       1                 yanleigis01
leigis02                       2                 yanleigis02
leigis03                       3                 yanleigis03
leigis04                       4                 yanleigis04
leigis05                       5                 yanleigis05
leigis06                       6                 yanleigis06
leigis07                       7                 yanleigis07
leigis08                       8                 yanleigis08
leigis09                       9                 yanleigis09
leigis10                       10               yanleigis10
leigis11                       11               yanleigis11

11 rows selected

由此可见，Oracle在创建函数索引的时候，实际上在表上添加了一个字段，用于保留该函数的计算结果，并基于此字段创建普通索引，
由于在底层数据字典中，将这个字段标记为 HIDDEN，因此用户在执行 DESC 的时候看不到，但可以用 SELECT 来查看。
在我们查找字段时，尽可能使用.user_tab_columns 而不是USER_TAB_COLS

因为USER_TAB_COLS可能加入oracle自己的隐含字段

于是，随之而来的问题就是Oracle是如何维护这个隐藏字段的呢？

SELECT COLUMN_NAME,COLUMN_ID,DATA_DEFAULT,HIDDEN_COLUMN,VIRTUAL_COLUMN,SEGMENT_COLUMN_ID,INTERNAL_COLUMN_ID
    FROM User_TAB_COLS
 WHERE TABLE_NAME='YL';
COLUMN_NAME    COLUMN_ID    DATA_DEFAULT    HIDDEN_COLUMN    VIRTUAL_COLUMN    SEGMENT_COLUMN_ID    INTERNAL_COLUMN_ID
------------------------   ------------------   ------------------------   ---------------------------   -----------------------------  ----------------------------------   -----------------------------------
ID                                      1                                                                   NO                                 NO                             1                                             1
NAME                                2                                                                  NO                                 NO                             2                                             2
SYS_NC00003$                              SUBSTR("NAME",4)               YES                                YES                                                                          3
可以看出，
1、普通的字段其 COLUMN_ID、SEGMENT_COLUMN_ID 是完全一致的，而Oracle自行添加的字段这两项却是空的，这个实际上是由于DBA_TAB_COLS的定义中做了转换：
decode(c.col#, 0, to_number(null), c.col#) as COLUMN_ID
decode(c.segcol#, 0, to_number(null), c.segcol#) as SEGMENT_COLUMN_ID(这里的c就是 sys.col$)，也就是说，实际上Oracle自行添加的字段这两项也是一致的，只不过是0而已
2、普通字段当然不会是隐藏、虚拟的，而Oracle自行添加的其 HIDDEN_COLUMN 和VIRTUAL_COLUMN 都是YES
3、INTERNAL_COLUMN_ID 完全与字段加入表中的顺序一致
4、注意Oracle自行添加的那个字段，将函数的表达式作为默认值

参考:http://www.itpub.net/thread-575391-1-1.html

来源：http://blog.csdn.net/yanleigis/article/details/2916096