Surety is performing system maintenance this weekend. Electronic date stamps on new Prior Art Database disclosures may be delayed.
Browse Prior Art Database

Build-in Translation in RDBMS

IP.com Disclosure Number: IPCOM000194982D
Publication Date: 2010-Apr-19
Document File: 1 page(s) / 16K

Publishing Venue

The IP.com Prior Art Database


Problem: One piece of data could have different presentation to different user or role. So, presenting data from RDBMS requires lots of meaning convert. The convert asks for lots of duplicate and low value codes. When a programmer develops an application presenting the data from a RDBMS, he/she need to translate the data in tables to meaningful contents to customers. An application will list the machines in server room to users. Administrators are interested in the IP or hostname of the machines. While business departments are interested in the business name of the servers. So for a same machine, it will be “svr001.cn.abc.com” to administrators, and will be “e-banking server A” to business department users. Know solution to this problem: Now, developers need to write code to perform the translation in their program. One can also use join-table query in database to get the result. Drawback of known solution: 1. Such kind of data translation is very popular in RDBMS applications. Using the known solution to implement, there will be a lot of duplicate and low value codes. Lots of effort is wasted. 2. If the data volume is huge, there will be performance issue using the current solution to solve this problem. The join table select between large tables may lead to performance 3. Another big issue is: if the value mapping changed, the code needs to be re-written. In the above example, if another department want to use another name for the machine, code must be re-written.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 91% of the total text.

Page 1 of 1

-in Translation in RDBMS

in Translation in RDBMS

To enable a DBMS for dynamic translation, we need to do the following things:
1. Provide a new kind of object, column map, which can be define on a mapping of a column.
2. Extend the Data Definition Language, to enable the creating of the new object: column map
3. Enable DBMS to accept a new parameter: map, during establishing a database connection. And store the map value into the database connection context. DBMS query engine will use the value of "map" to generate query result.

4. Enable DBMS to query and return the map value for a column if it has a map object, using the "map" value in the current connection context. Important, the query optimizer should generate query plan for such kind of query automatically.

For example, DBMS provide a new object, could be named as column map. The map object contains the values mapping to the translation types. For example, Table Companies {
serverid char(8),
servername varchar(20)

DBMS users can define a column map on column compname as below: type map value
0 server01.cn.abc.com
1 e-banking server A
2 accounting server

DBMS connection accepts a new connect parameter in database connect string, for example:


Client program connect to DBMS using this connect string; DMBS will store the map value in the currentconnection context. In this connection context, when client program submit the SQL statement of : "select serverid, servername from Compan...