Browse Prior Art Database

Method of Macthing UI Translation from PII for Multilingual GUI Automation Testing

IP.com Disclosure Number: IPCOM000238919D
Publication Date: 2014-Sep-25

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a method to macth the values of variables that are used in automation scripts through extracting Program Integrated Information(PII, refer to resource bundle files) for multilingual GUI automation testing, in order to reuse English automation scripts.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 50% of the total text.

Page 01 of 11

Method of Macthing UI Translation from PII for Multilingual GUI Automation Testing

Nowadays, more and more software released with built-in multiple language support. Automation scripts are usually based on English build. There are the following problems when reuse English automation scripts to multiple languages:

High automation failure rate because of untranslated string in automation script.


1.

Automation scripts usually recognize UI object by the attribute of the object. Undeniably, it is a best way to use a fixed attributes


2.

that is language independent, such as ID. However, in our experience, most of object's ID(.id) properties are changed when we change builds in the development cycle. On the other side, an important focus of multilingual automation is to verify the translated UI string, so the fixed attributes (such as.id) is not good choice for multilingual automation script. Most of time, the text label is used as the attribute to identify the object.

Maintaining a set of automation scripts for each language is not an adequate option, as it is time-consuming. The more


3.

automation scripts, the more effort require.

In our experience, most automation framework is based on script programming instead of record-replay mode, so the existing


4.

method that erases language specific information from object repositories is not very effective for such automation framework.

Based on the analysis above, the actual situation is that most existing automation framework identifies object by the attribute of text label. And the biggest challenge is how to get translated language specific information for the variables defined in automation scripts when execute English automation scripts with multilingual environments.

This disclosure discloses a method to transform the values of variables that are used in automation scripts through extracting translated language resource files.

The core ideas is to generate a pseudo build with a static unique ID added to resource bundle files' each value, have a new value (UI

1


Page 02 of 11

text label$Definition file Name$Variable Name#) assigned to variables that are defined in automation scripts to represent UI text label, and then run English automation scripts on pseudo build to setup the mapping between variables usedin automation scripts and resource bundle files' key. Based on mapping files, translated string can be extracted easily from language resource bundle files for each variables used in automation scripts

The method includes five steps:
1. Add one unique id for each translatable item in PII file. (For example, one UI string is displayed as
"New"in English build and the UI will be displayed as"New$TASBI#"in pseudo build. The unique id $TASBI#is stead for the full path of the string stored in PII resource file)


2. Kit a pseudo build base on the PII files processed in step #1.


3. Add an identifier to each variable in automationscripts (For example, English value of sNewvariable is "New"inI...