CN101488148A - Data compilation method in multiple databases - Google Patents

Abstract

The invention provides a method of arranging and collecting data from a plurality of databases. In the method, script description is used, isomorphic data arrangement is unitized, and heterogeneous data processing is specialized; a plurality of databases are numbered, rules are set in the script, own techniques of the databases are utilized simultaneously to complete the integrated processing of the data, and the real-time collection or timing collection of the data is finished through a real-time calling interface or through the binding with the timing event processing; the invention works during the report development of multi-database or the data intensive collection, and when in use, the method carries out the task description by a simple label script and supports the real-time or timing collection of the data from a plurality of databases. The method of arranging and collecting the data from a plurality of databases has the advantages of providing API finished data results for use, realizing the data template extraction of isomorphic databases, improving the developing efficiency and collecting the isomorphic and heterogeneous databases according to corresponding rules.

Description

A kind ofly from a plurality of databases, carry out the method that data preparation collects

Technical field

The present invention relates to a kind of Computer Applied Technology, use, can realize a plurality of data of database are put in order the technology that collects, develop the technical support that provides the foundation for form exploitation and data gather towards the professional and technical personnel.

Background technology

Business data can relate to numerous databases at present, the data in the associated databases is put in order, is collected need pay a large amount of time and cost.Present technique is distinguished purpose data classifying the data of isomorphism or isomery simultaneously by setting up a kind of data preparation flow process of standard, the isomorphism data of database is collected carry out templating, thereby reduce cost of development, improves the quality of exploitation.

Summary of the invention

The purpose of this invention is to provide a kind of method that data preparation collects of from a plurality of databases, carrying out.

The objective of the invention is to realize in the following manner, step is as follows:

1) use script describing, the data preparation of isomorphism is unitized, and the processing of isomeric data is becomed privileged;

2) a plurality of databases are numbered, and in script, set rule, the technology of utilizing database itself to carry is simultaneously finished the integrated processing of data, and handle by the real-time calling interface or with timed events and to bind, finish collecting in real time or regularly collecting of data;

3) play a role when the form exploitation of striding multiple database or data centralization gather, in use, carry out task description by the simple label script, concrete label script is as follows:

/ *---database configuration joint, for example: 0001..0002,0004,0006---*/--use comma to do connector between two accounts connection configuration numberings, if use ' .. ' connector, expression from XX number to XX number, represent 0001 such as 0001..0004,0002,0003,0004

<DATABASE>

</DATABASE>

/ *--result set configuration section---*/

The table name that/* 00 result deposits, must be with WDT_ beginning table */--need in current accounts, create a such entity list, wherein must comprise database numbering and database-name field, these two field systems are defaulted as respectively: DB_ID, if DB_NAME. do not adopt default value, build then that two field names of table correspondence must be with user-defined consistent as a result.

<RESULT_TABLE>

</RESULT_TABLE>

The condition of the data of/* 01 deletion respective table, NONE=do not carry out deletion */

<RESULT_DELETE_WHERE>

</RESULT_DELETE_WHERE>

/ * 02 is used to indicate the order of the deposit data row that collected out, the SQL of this order and last acquisition data be consistent */

<RESULT_COLUMN>

</RESULT_COLUMN>

/ * 03 is the field of expression database numbering in the table as a result, do not write the Cai Yong Summerside value of recognizing DB_ID*/

<RESULT_DB_ID_COLUMN>

DB_ID

</RESULT_DB_ID_COLUMN>

/ * 04 is the field of expression database-name in the table as a result, do not write the Cai Yong Summerside value of recognizing DB_NAME*/

<RESULT_DB_NAME_COLUMN>

DB_NAME

</RESULT_DB_NAME_COLUMN>

/ *---purpose data classifying process can be SQL, also can be the execution storing process, if increase step Sudden, uses the numbering area separately just passable---*/

<STEP1>

</STEP1>

<STEP2>

</STEP2>

/ *---result set obtains statement, can be SQL, also can be the execution storing process, but will notice that the order that is listed as can carry out corresponding with the result set row of front---*/

<RESULT_SQL>

</RESULT_SQL>

Function that has and beneficial effect

2. support from a plurality of databases, data to be carried out in real time or regularly collect gathering.

3. provide API to finish the use of data result.

4. can realize the templating of isomorphism data of database is extracted, improve development efficiency.

5. the database of isomery and isomorphism can be concentrated according to corresponding rule.

Embodiment

Configuration database connects, and finishes each connection that need collect database and points to.

Set up the middle table that the purpose data classifying script needs, contents such as storing process, configuration data collects script then.

In the form computing, call, or in timing services, call.

Script uses step as follows:

Function: from four different databases, collect department table:

The sql of the table as a result statement of being built:

CREATE?TABLE?WDT_RESULT

(

BMBH?VARCHAR(20)NULL，

BMMC?VARCHAR(60)NULL，

DB_ID?VARCHAR(20)NULL，

DB_NAME?VARCHAR(60)NULL

)

go

The purpose data classifying definitional part:

/ *---database configuration joint, for example: 0001..0002,0004,0006---*/

<DATABASE>

0001，0002，0003，0004

</DATABASE>

/ *--result set configuration section---*/

The table name that/* 00 result deposits, must be with WDT_ beginning table */

<RESULT_TABLE>

WDT_RESULT

</RESULT_TABLE>

The condition of the data of/* 01 deletion respective table, NONE=do not carry out deletion */

<RESULT_DELETE_WHERE>

</RESULT_DELETE_WHERE>

/ * 02 is used to indicate the order of the deposit data row that collected out, the SQL of this order and last acquisition data be consistent */

<RESULT_COLUMN>

BMBH，BMMC

</RESULT_COLUMN>

/ * 03 is the field of expression database numbering in the table as a result, do not write the Cai Yong Summerside value of recognizing DB_ID*/

<RESULT_DB_ID_COLUMN>

DB_ID

</RESULT_DB_ID_COLUMN>

/ * 04 is the field of expression database-name in the table as a result, do not write the Cai Yong Summerside value of recognizing DB_NAME*/

<RESULT_DB_NAME_COLUMN>

DB_NAME

</RESULT_DB_NAME_COLUMN>

/ *---result set obtains statement, can be SQL, also can be the execution storing process, but will note the order that is listed as

Can carry out corresponding with the result set row of front---*/

<RESULT_SQL>

SELECT?KCBMZD_BMBH，KCBMZD_BMMC?FROM?KCBMZD?GROUP?BY

KCBMZD_BMBH，KCBMZD_BMMC

</RESULT_SQL>

The main program assembly:

■ purpose data classifying processing components: n_cst_wd_multi_database

DLL (dynamic link library) method: public function integer of_exec_pre (string as_id, string

As_arg[], string as_data[]), the processing that is used to finish some purpose data classifying processes is called.

The data that ■ is relevant are built the table order:

Database connects allocation list

CREATE?TABLE?WDCXDB(

F_ID VARCHAR (10) NOT NULL ,/* be used for unique identification corresponding database numbering */

F_NAME VARCHAR (100) NOT NULL, it is database where that/* is used to indicate, and refers generally to

The title * of bright factory/

F_TYPE VARCHAR (20) NOT NULL ,/* is used to indicate the type of database: ORACLE/

SYC/MSS/ODBC*/

F_SERVER VARCHAR (60) NOT NULL ,/* are used to indicate the server of database at current machine

Connection configuration * on the device/

F_DATABASE VARCHAR (60) NOT NULL ,/* be used to indicate be which database */

F_LOGID VARCHAR (60) NOT NULL ,/* login people */

F_PASSWORD VARCHAR (30) NULL ,/* login password */

F_BZ VARCHAR(200) NULL)

go

CREATE?UNIQUE?INDEX?I_WDCXDB?ON?WDCXDB(F_ID)

go

The purpose data classifying allocation list, the Script section is preserved by the data stream storage format.

CREATE?TABLE?WDCXPRE(

F_ID VARCHAR (10) NOT NULL, the numbering * of/* pre-treatment/

F_NAME VARCHAR (60) NOT NULL, the title * of/* pre-treatment/

F_AUTORUN CHAR (1) default ' 0 ' null ,/* whether move automatically */

F_RUNDIF u001 default 0 null, the interval of/* operation, unit be second a */

F_RUNLAST u001 default 0 null, the time of the last operation of/*, unit is

Second, rise start from */

F_RUNNEXT u001 default 0 null ,/* need next time the time * that moves/

F_BZ VARCHAR (200) null ,/* remarks */

F_LOCK CHAR (1) NULL ,/* whether locked */

F_CONT IMAGE NULL)/and the concrete content of *, use scale-of-two

The storage, but make be editing machine edit */

go

create?unique?index?I_WDCXPRE?on?WDCXPRE(F_ID)

go

Embodiment:

■ calls the report making that this technology is carried out the integration across database system in reporting system,

N_cst_jxc_pub_cx.of_pre_sql () invoke code is as follows:

...

// if carry out the integration across database purpose data classifying, so just carry out pre-treatment and return then.

if?left(upper(trim(ls_sql))，2)＝’P:’then

n_cst_wd_multi_database?lnv_mul

lnv_mul＝create?n_cst_wd_multi_database

get_token(ls_sql，’:’)

li_ret＝lnv_mul.of_exec_Pre(ls_sql，is_sjlm，is_value)

destroy?lnv_mul

return?li_ret

end?if

….

■ calls this technology in timing services n_cst_msg_server service object:

■ instance variable: n_cst_wd_multi_database inv_mul//be used to the carry out purpose data classifying of many accounts

Handle

Invoke script part among the ■ method of_gen_message ():

......

Case ' PR ' // when for pre-treatment, call the mode of pre-treatment.

IF?INV_MUL.OF_EXEC_PRE(LS_COL，ls_p1，ls_p2)＝1THEN

Sss=ls_pre+ ' carries out many accounts purpose data classifying and handles ['+ls_col+ '] success ~ r ~ n '

else

Sss=ls_pre+ ' carries out many accounts purpose data classifying and handles ['+ls_col+ '] failure: '+

inv_mul.is_error+’~r~n’

end?if

……。

Claims (1)

1. one kind is carried out the method that data preparation collects from a plurality of databases, it is characterized in that step is as follows:

1) use script describing, the data preparation of isomorphism is unitized, and the processing of isomeric data is becomed privileged;

2) a plurality of databases are numbered, and in script, set rule, the technology of utilizing database itself to carry is simultaneously finished the integrated processing of data, and handle by the real-time calling interface or with timed events and to bind, finish collecting in real time or regularly collecting of data;

3) play a role when the form exploitation of striding multiple database or data centralization gather, in use, carry out task description by the simple label script, concrete label script is as follows:

/ *---database configuration joint, for example: 0001..0002,0004,0006---*/--use comma to do connector between two accounts connection configuration numberings, if use ' .. ' connector, expression from XX number to XX number, represent 0001 such as 0001..0004,0002,0003,0004

<DATABASE>

</DATABASE>

/ *--result set configuration section---*/

The table name that/* 00 result deposits, must be with WDT_ beginning table */--need in current accounts, create a such entity list, wherein must comprise database numbering and database-name field, these two field systems are defaulted as respectively: DB_ID, if DB_NAME. do not adopt default value, build then that two field names of table correspondence must be with user-defined consistent as a result.

<RESULT_TABLE>

</RESULT_TABLE>

The condition of the data of/* 01 deletion respective table, NONE=do not carry out deletion */

<RESULT_DELETE_WHERE>

</RESULT_DELETE_WHERE>

/ * 02 is used to indicate the order of the deposit data row that collected out, the SQL of this order and last acquisition data be consistent */

<RESULT_COLUMN>

</RESULT_COLUMN>

/ * 03 is the field of expression database numbering in the table as a result, do not write adopt default value DB_ID*/

<RESULT_DB_ID_COLUMN>

DB_ID

</RESULT_DB_ID_COLUMN>

/ * 04 is the field of expression database-name in the table as a result, do not write adopt default value DB_NAME*/

<RESULT_DB_NAME_COLUMN>

DB_NAME

</RESULT_DB_NAME_COLUMN>

/ *---purpose data classifying process can be SQL, also can be the execution storing process, if increase step Sudden, uses the numbering area separately just passable---*/

<STEP1>

</STEP1>

<STEP2>

</STEP2>

/ *---result set obtains statement, can be SQL, also can be the execution storing process, but will notice that the order that is listed as can carry out corresponding with the result set row of front---*/

<RESULT_SQL>

</RESULT_SQL>。