Question

我在excel中使用coulmns有一些数据，我需要将这些数据从excel迁移到Oracle中的某些表。我如何以最佳和最快的方式实现这一目标。我需要在Oracle中从该单个excel文件中加载5个表。任何可用于此类任务的工具？

Answer 1

An approach could be the following.

Say you have two tables like these

create table tabA (col_a number, col_b number);
create table tabB (col_b number, col_c varchar2(10));

and you need to load data starting from a a file d:\file.csv like this:

col_a,col_b,col_c
10,99,"a"
20,999,"b c d"

you first need to create a directory, then build a table containing the CSV:

CREATE DIRECTORY dir_d AS 'D:\';
CREATE TABLE csvFile
(
    Col_a                                      VARCHAR2(4000),
    Col_b                                      VARCHAR2(4000),
    Col_c                                      VARCHAR2(4000)
)
ORGANIZATION EXTERNAL
    (TYPE ORACLE_LOADER
          DEFAULT DIRECTORY dir_d
              ACCESS PARAMETERS
       (RECORDS DELIMITED BY NEWLINE
        SKIP 1
        FIELDS
            TERMINATED BY ','
            OPTIONALLY ENCLOSED BY '"'
            LRTRIM
            MISSING FIELD VALUES ARE NULL
      )
          LOCATION('file.csv'));

Now you can use SQL to insert data in your tables starting from the CSV table, using every logic you need; for example:

insert into tabA select col_a, col_b from csvFile;
insert into tabB select col_b, col_c from csvFile;

SQL> select * from tabA;

     COL_A      COL_B
---------- ----------
        10         99
        20        999

SQL> select * from tabB;

     COL_B COL_C
---------- ----------
        99 a
       999 b c d

Answer 2

如果您只想将数据放入一个表中，可以使用sqldeveloper将数据导入表中，如显示此tutorial：

Answer 3

Oracle的SQL*Loader工具可以处理多表插入。如果您不想在构建适当的控制文件时遇到麻烦，可以先使用您选择的工具将数据加载到临时表中。

然后，您可以从登台表执行单个插入，或使用conditional（？）muti table insert语句。：

insert all into table_a (cola, colb, colc) values (col_a, col_b, col_c)
           into table_b (cola, colb, colc, cold) values (col_a, col_b, col_c col_d)
  select col_a, col_b, col_c, col_d from staging_table;

或

insert all 
  when col_a < 1000 then
    into table_a (cola, colb, colc) values (col_a, col_b, col_c)
  when col_a between 900 and 10000 then
    into table_b (cola, colb, colc, cold) values (col_a, col_b, col_c col_d)
  else
    into table_c (cola, colb, colc, cold) values (col_a, col_b, col_c col_d)
  select col_a, col_b, col_c, col_d from staging_table;

使用条件插入，每个WHEN子句的计算结果为true将导致数据被插入到指定的表中，而ELSE子句仅在没有其他选项适用时才会起作用。

这是ETL工作的最佳方式？

3 个答案: