Testing iOS Apps with HadoopUnit: Rapid Distributed GUI Testing by Scott Tilley & Krissada Dechokul

Author:Scott Tilley & Krissada Dechokul [Tilley, Scott]
Language: eng
Format: azw3
ISBN: 9781627056526
Publisher: Morgan & Claypool Publishers
Published: 2015-07-15T16:00:00+00:00

Figure 3.2: Sample of a test case list.

A test case name is used as a key to uniquely identify each test execution by the Map function. It is used as a key to merge together test results into one report by the Reduce function. There can be multiple executions of different commands for a single test case name, meaning that there can be multiple lines in the list with the same test case name. In this case, after finish executing all the test cases, each test execution will be merged into one line by the Reduce function. When a test case fails, this test case name is used as a name of directory to store the trace document in the HDFS.

A test execution command is the directive to execute a test case. The command will be executed as an external process by the Map function of HadoopUnit. This component can be in a form of a simple Instruments command or any type of executable command. The execution commands demonstrated in Figure 3.2 are in a form of executable Rake tasks receiving a path to a UI Automation test script. More information about Rake is explained in the next section.

The location of the trace document is an extra component that is added to each line of text in this version of HadoopUnit to accommodate the way Instruments treats the test results of each test case execution. This is the directory that Instruments uses to store a trace document for the test execution. When a test case fails, HadoopUnit uses this information to retrieve the trace document of the test execution and copy it into the HDFS.

The original version of HadoopUnit utilizes Ant, a build tool for Java projects, to extract test cases to be executed from a test suite, as well as other information needed to run the test cases such as necessary classpaths or libraries required to build and run a particular project. This test case list, however, can be constructed manually using a simple text editor or using a simple script to generate the list programmatically; no special tool is necessary.

Once the test case list has been created, it needs to be uploaded to the HDFS. Its path in the HDFS will be used as one of a parameter in the command to initiate a Hadoop job. The command to initiate a Hadoop job will be explained in detail in the following sections.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.