Browse Prior Art Database

Optimizing Host Application Presentation Space Recognition Events Through Heuristic Comparison

IP.com Disclosure Number: IPCOM000013779D
Original Publication Date: 2000-Dec-01
Included in the Prior Art Database: 2003-Jun-18
Document File: 2 page(s) / 189K

Publishing Venue

IBM

Abstract

This disclosure describes a method and apparatus for improving the efficiency of legacy host application screen recognition. Refer to the Host Application Presentation Space Recognition Producing Asynchronous Events disclosure (not in patent database, too old?), which essentially describes an API apparatus that generates recognition events when a screen matching a description provided by the user appears in the apparatus' presentation space. If a screen is matched, the user's program code is called and he may handle it anyway he chooses (scrape text, send keys, etc.). The system stores all the user's screen descriptions internally in a vector object and just iterates through the vector for every screen sent from the host. This iterative description comparison process presents a problem for managing very large amounts of screen descriptions and was the spark for the current disclosure. Say for instance there is a user like Ford Motor Company, Inc. that has used legacy host systems for decades and has built a huge host application base with thousands of screens. The Ford developer wants to code descriptions for every screen and give them to the screen recognition apparatus, thus adding thousands of descriptions that need to be compared. Currently, the recognition apparatus will loop through every item in the vector of descriptions until it sees a match. If a frequently occurring screen was added at number 500 there will be a lot of comparison failures before the screen is hit: poor performance. The current disclosure solves this problem by gathering screen descriptions based on heuristics that make for efficient screen comparison when the screen recognition system uses them. The heuristics used are actually quite straightforward. The idea is that we want to hit the screen descriptions that are the fastest to compare first. Below is a ranking in speed for each type of descriptor that makes up a description (fastest to slowest): 1. Operator Information Area status (is the keyboard locked)

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 2

  Optimizing Host Application Presentation Space Recognition Events Through Heuristic Comparison

This disclosure describes a method and apparatus for improving the efficiency of legacy host application screen recognition. Refer to the Host Application Presentation Space Recognition Producing Asynchronous Events disclosure (not in patent database, too old?), which essentially describes an API apparatus that generates recognition events when a screen matching a description provided by the user appears in the apparatus' presentation space. If a screen is matched, the user's program code is called and he may handle it anyway he chooses (scrape text, send keys, etc.). The system stores all the user's screen descriptions internally in a vector object and just iterates through the vector for every screen sent from the host.

This iterative description comparison process presents a problem for managing very large amounts of screen descriptions and was the spark for the current disclosure. Say for instance there is a user like Ford Motor Company, Inc. that has used legacy host systems for decades and has built a huge host application base with thousands of screens. The Ford developer wants to code descriptions for every screen and give them to the screen recognition apparatus, thus adding thousands of descriptions that need to be compared. Currently, the recognition apparatus will loop through every item in the vector of descriptions until it sees a match. If a frequently occurring screen was added at number 500 there will be a lot of comparison failures before the screen is hit: poor performance.

The current disclosure solves this problem by gathering screen descriptions based on heuristics that make for efficient...