Browse Prior Art Database

A method to generate object maps and actions from application source codes

IP.com Disclosure Number: IPCOM000220952D
Publication Date: 2012-Aug-16
Document File: 9 page(s) / 202K

Publishing Venue

The IP.com Prior Art Database

Abstract

In GUI(Graphic User Interface) automation testing, we usually create object maps to list the test objects in the application-under-test.A commonly known way to generate the object map is to launch the application (or open the url in a browser), then use automation tool to capture the objects(buttons, checkboxes, etc) from the GUI and add them to object map.However, to capture the objects from the GUI and add them to object map can be a time-consuming work. And you will need to launch and exercise the application to make the objects appear on the GUI(you cannot capture the objects that do not exist on the GUI), if there is some blocking defects, you can not navigate to the subsequent GUI and capture the objects.Our invention provides a way to generate object maps for GUI automation test quickly by analyzing source codes of application-under-test. For example, for C++ application, most of the GUI components(dialogs, buttons, check boxes, etc) are defined in resource files (.rc). For the Web application, the GUI objects are defined in the HTML or JS file. So we can extract and analyze the GUI objects from the resource file (.rc or html), and then generate the object maps and simple action methods.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 53% of the total text.

Page 01 of 9

A method to generate object maps and actions from application source codes

Our invention provides a way to generate object maps for GUI automation test quickly by analyzing source codes of application-under-test.

In GUI(Graphic User Interface) automation testing, we usually create object maps to list the test objects in the

application-under-test. It is a static view that describes test objects known to Functional Test in the application-under-test. The test object map contains recognition properties for each object, you can update the information in one location. Any scripts that

reference that test object map share the updated information .

Commonly known solution:

A commonly known way to generate the object map is to launch the application(or open the url in a browser), then use automation tool to capture the objects(labels, buttons, checkboxes, text fields, etc) from the GUI and add them to object map.

Drawbacks of the known solution:

However, to capture the objects from the GUI and add them to object map can be a time-consuming work. And you will need to

launch and exercise the application to make the objects appear on the GUI(you cannot capture the objects that do not exist on the GUI), if there is some blocking defects, you can not navigate to the subsequent GUI and capture the objects.

Our invention provides a way to generate object maps for GUI automation test quickly by analyzing source codes of application-under-test. For example, for C++ application, most of the GUI components(dialogs, buttons, check boxes, radio buttons, dropdown lists, etc) are defined in resource files (.rc), we may generate corresponding object maps by analyzing the

resource files.

Advantage(s) of the our invention:


1. Testers do not need to execise the application to capture the GUI components when developing automation scripts .

1


Page 02 of 9


2. Testers can start developing automation scripts as soon as the GUI definition is available, they don't need to wait until the application is ok to run.


3. The invention may sharply reduce the time that testers spend on finding and defining GUI objects .


4. The invention may quickly adapts the object maps to GUI changes.

All GUI objects must be defined somewhere in the source codes. For C++ application, all GUI objects are defined in the resource files (.rc files). It includes all dialogs and GUI objects in the dialogs. For the Web application, the GUI objects are defined in the HTML file. So we can extract and analyze the GUI objects from the resource file (.rc or html), and the...