Some applications, such as Lawson Portal or Oracle PeopleSoft, are delivered via a web browser. Depending on how the webpage is setup to render, the content of that page can shift around (tabs and bars being added or moved is a common cause, or the end user’s permissions to view content). This causes an issue when ImageNow / Perceptive Content is attempting to learn values from pages such as this.
The best way to address these “shifting values” is to add custom HTML tags to the web pages. Typically, these hidden HTML tags are not affected by an upgrade of the host application. This does require, however, that you have access to the applications source code. Some applications, such as PeopleSoft, have a tool for this (peopletools).
Lets take a look at what a typical hidden HTML field looks like:
The “name” parameter should be a static identifier, which we will call with the LearnMode. The value should be dynamic, and come from the host application (although, if this is not possible, we can still use the hidden field as an anchor). ImageNow / Perceptive Content can now learn the screen, using the hidden input type. Look for the name (“studentID” in the above example) and pull the value from that hidden HTML field into the application plan. With this approach, it does not matter if the data shifts position on the screen. ImageNow will always look for an HTML element with that name and grab its value.
Also note, that this method can be used to identify the screen within the application plan as an anchor. For instance, this line of code allows LearnMode to use the value “Transcript Screen” as the screen identifier.
<input type=”hidden” name=”SCREEN_NAME” value=”Transcript Screen”>