Dynamic Policy Based Host Screen Data Extraction
Original Publication Date: 2000-Jan-01
Included in the Prior Art Database: 2003-Jun-19
Method and apparatus for policy-based data extraction from legacy host system datastream screens (3270,5250,VT). Using the current invention, a developer displays different amounts or "levels' of data from host screens based on a policy lookup in a policy directory. Given a user's policy level, more or less data from a given host screen is shown to the user. One possible implementation of the invention would be to develop an “Extract Policy Engine” using IBM’s Host On-Demand Screen API’s in conjunction with a policy directory service. First, the engine would use the policy directory to look up and validate the user for the extract. Once the user’s data privileges are determined it is known what data can be extracted from the host screen. Next the engine would extract the applicable data from the screen using HOD based on the user’s privileges. Like previous work in the screen request area, users receive screens either through their initiation or asynchronously from the host. Either way, the policy engine filters the data from the screen extractions. Addressing the concerns about the similarity between the current patent and the SmartMaster portion of the patent disclosure filed by Li, Tan, and Webb (RSW8-1999-0087), the current disclosure describes dynamically limiting data extracted from screens. Although disclosure RSW8-1999-0087 essentially limits the data provided to the user by using XSL, which can be retrieved from a policy directory, the extract used by the XSL SmartMaster disclosure is static. This means that only one screen data extraction happens for all users; therefore, all the data from that extract is passed to the XSL engine. Even if the XSL for a given user only needs one data field on the screen, all the data fields for all the users are extracted every time.