This application relates generally to systems, methods and apparatuses, including computer programs, for testing mobile applications. More specifically, this application relates to using simulated human behaviors during mobile application testing to identify, track and resolve technical issues occurring during mobile application usage.
Mobile automation testing is currently used to identify and resolve bugs and other technical issues with mobile applications, thereby ensuring that mobile applications achieve high performance and stability. Traditional mobile automation testing includes testing mobile application features or functions one at a time, e.g., by writing one or more test cases. However, even if all test cases are passed, it is difficult to ensure that the mobile application contains no bugs or coding errors, because the test cases are unable to take human behaviors into account. For example, random events such as interruptions caused by notifications received on a phone, incoming calls, and switching applications by the user can give rise to additional bugs and errors within the application that are outside the scope of traditional mobile automation testing. Such real-life randomness injects significant uncertainty into mobile application testing.
To compound these problems, once an error is discovered in a real usage scenario, it is very hard to reproduce the error, particularly if it is caused by a random event during testing. Examples of such scenarios include: (1) a phone call is received while the application is running; after the user answers the phone, the application's status is corrupted, or the application fails to respond; (2) a notification pops up while the application is running; after the user reads the message by clicking the notification, the application crashes; (3) a new application is started using the test application; after the user closes the new application, the test application does not display correctly; (4) a mobile device is moved from one position to another randomly, e.g., while a user gestures in conversation, so the application switches from landscape display mode to portrait display mode; these random movements give rise to bugs; (5) a network to which the mobile device is connected switches one or off randomly, e.g., loses an internet connection or switches WiFi connections; these random network status changes give rise to bugs.
Accordingly, the invention provides systems, methods and apparatuses to test mobile applications by emulating human or “human-like” behaviors in simulated scenarios that better approximate or mimic real-life scenarios. A framework is designed that can be integrated with current mobile automation test environments and can generate random events (e.g., switching applications, receiving push notifications, making phone calls, moving or rotating the device, etc.) while executing normal test scripts on the mobile testing device to emulate human-like behaviors. The invention enables developers to reproduce defects, which in turn allows them to gain a better handle on ways to fix the defects. This setup stands in contrast to normal mobile application testing, in which a defect is difficult to reproduce once it is discovered and/or tested only once. Thus, the invention enables high-quality testing, tracking and resolution of a variety of previously undiscovered and/or unresolved technical issues during mobile automation testing.
In one aspect, the invention features a computerized method of testing a mobile application stored in memory of a mobile test device. The computerized method includes executing, by a computing device, a test script stored in memory of the computing device. The computerized method also includes running, by the computing device, according to the test script, a human behavior simulation module stored in memory of the computing device. The computerized method also includes retrieving, by the computing device, via the human behavior simulation module, an event list stored in memory of a server. The server is in electronic communication with the computing device. The event list includes one or more electronic human behavior simulations. The computerized method also includes receiving, by the computing device, via the human behavior simulation module, an electronic human behavior simulation from the event list. The computerized method also includes providing, by the computing device, via the human behavior simulation module, selected data of the electronic human behavior simulation to memory of the server for recordation. The computerized method also includes executing, by the computing device, via the human behavior simulation module, the electronic human behavior simulation on a mobile test device in electronic communication with the computing device. The computerized method also includes executing, by the computing device, according to the test script, a test case stored in memory of the computing device on the mobile test device.
In some embodiments, the electronic human behavior simulation is received at a random time. In some embodiments, the electronic human behavior simulation is generated based on a record of a previously run electronic human behavior simulation. In some embodiments, the record is stored in memory of the server and including at least one parameter of the previously run electronic human behavior simulation. In some embodiments, executing the electronic human behavior simulation and executing the test case occur simultaneously or near-simultaneously. In some embodiments, the electronic human behavior simulation simulates a human making a phone call. In some embodiments, the electronic human behavior simulation simulates a human receiving a phone call. In some embodiments, the electronic human behavior simulation simulates a human rotating or translating a mobile device. In some embodiments, the electronic human behavior simulation simulates a human switching between mobile applications. In some embodiments, the electronic human behavior simulation simulates a human rotating a mobile device such that a display screen of the mobile device changes orientation. In some embodiments, the electronic human behavior simulation simulates changing a network connectivity status of a mobile device. In some embodiments, the electronic human behavior simulation simulates receiving a push notification on a mobile device.
In some embodiments, the method includes logging, by the computing device, via a log module in communication with the human behavior simulation module, the selected data of the electronic human behavior simulation in a database of the server. In some embodiments, executing the electronic human behavior simulation includes providing, via the human behavior simulation module, the electronic human behavior simulation to an event execution module. In some embodiments, the event execution module is at least one of a command execution module, (e.g., including WebDriverAgent/UIAutomator), a notification module, or an internet-of-things module. In some embodiments, executing the electronic human behavior simulation includes sending a command to mobile device test equipment in electronic communication with the computing device.
In some embodiments, the mobile device test equipment is configured to interact with the mobile test device. In some embodiments, the electronic human behavior simulation includes a physical movement implemented by auxiliary hardware in electronic communication with the mobile device test equipment. In some embodiments, the mobile device test equipment includes a mobile device test fixture. In some embodiments, the mobile device test fixture is configured to rotate or translate the mobile test device or to interact with a screen of the mobile test device. In some embodiments, the mobile device test equipment includes an auxiliary mobile phone used to make a phone call to the mobile test device. In some embodiments, the electronic human behavior simulation includes providing a push notification to the mobile test device via a cloud messaging service in electronic communication with the mobile test device. In some embodiments, the test script includes instructions specifying a time at which to run the human behavior simulation module.
In another aspect, the invention features a system for testing a mobile application stored in memory of a mobile test device. The system includes a computing device configured to (i) execute a test script stored in memory of the computing device; (ii) run, according to the test script, a human behavior simulation module stored in memory of the computing device; (iii) retrieve, via the human behavior simulation module, an event list including one or more electronic human behavior simulations; (iv) receive, via the human behavior simulation module, an electronic human behavior simulation from the event list; (v) provide, via the human behavior simulation module, selected data of the electronic human behavior simulation to memory of the server for recording; (vi) execute, via the human behavior simulation module, the electronic human behavior simulation on a mobile test device in electronic communication with the computing device; and (vii) execute, according to the test script, a test case stored in memory of the computing device on the mobile test device. The system also includes a server in electronic communication with the computing device. The system also includes a mobile test device in communication with the computing device, the mobile test device configured to host the mobile application and execute mobile automation test scripts.
In some embodiments, the system further includes a cloud messaging service in electronic communication with the computing device. In some embodiments, the cloud messaging service is configured to push notifications to the mobile test device. In some embodiments, the system further includes mobile device test equipment in electronic communication with the computing device. In some embodiments, the mobile device test equipment includes at least one of a mobile device test fixture and an auxiliary mobile phone. In some embodiments, the human behavior simulation module includes a human behavior generator, a log module, and at least one of a command execution module, a notification module, or an internet-of-things module. In some embodiments, a command execution module is used to execute scripts on mobile devices. In some embodiments, WebDriverAgent and UIAutomator are two frameworks of command execution modules. In some embodiments, WebDriverAgent is used to execute scripts on iOS devices. In some embodiments, UIAutomator is used to execute scripts on Android devices.
By virtue of the above systems, methods, apparatuses, and those further described below, the invention provides a novel solution to enhance mobile automation testing. A framework is designed and used to generate human-like behaviors while running test scripts to simulate mobile application usage in real scenarios, such as making a phone call, receiving a push notification, or switching applications. In addition, when a bug is tested out, the system can reproduce the bug for developers to fix it.
The advantages of the invention described above, together with further advantages, may be better understood by referring to the following description taken in conjunction with the accompanying drawings. The drawings are not necessarily to scale; emphasis is instead generally placed upon illustrating the principles of the invention.
The human behavior simulation module 112 can provide at least three functions: (1) it interacts with the mobile test devices 124 in a human-like manner, e.g., by sending notifications, making phone calls, or switching applications or network connections; (2) it sends commands to the mobile device test equipment 120 to drive the mobile device test equipment 120 or move the mobile test devices 124; (3) it records and/or retrieves tested behaviors to and/or from the server 116. The server 116 then stores human-like behaviors and test logs, and human-like behaviors can be added, updated, deleted, and/or stored in the database. The mobile device test equipment 120, on which mobile test devices 124 having installed mobile applications 128 are fixed for testing, can receive commands from the human behavior simulation module 112 (e.g., a human behavior generator framework) to actuate certain movements of the mobile test devices 124, such as shifting and rotating. When a bug is tested, records can be retrieved from the server 116 and the test scripts can be replayed to show how the bug is generated.
The server 116 can include an event list 132, which can store one or more electronic human behavior simulations, and memory 136 for recordation (e.g., of testing events). The one or more test scripts 104 can be used to test features and/or functions of the one or more mobile applications 128 using one or more test cases. They can also call exposed methods from the human behavior simulation module 112 (e.g., automatically or randomly) in a manner that does not require testers to modify their existing test scripts to indicate when they should call random human-like events. In some embodiments, the system includes an “on/off” feature that allows the same test scripts to be used for their traditional automation testing (e.g., in system “off” mode) or by simulating a real-life usage scenario of mobile apps (e.g., in system “on” mode). In some embodiments, the system includes a cloud messaging service 140 in electronic communication with the computing device, the cloud messaging service 140 configured to push notifications to the one or more mobile test devices 124.
Using the system 100, the mobile application 128 can be tested according to a computerized method. The computing device 104 executes a test script 108 stored in memory of the computing device 104. The computing device 104 calls, according to the test script 108, the human behavior simulation module 112 stored in memory of the computing device 104. The computing device 104 retrieves, via the human behavior simulation module 112, the event list 132 stored in memory of the server 116, the event list 132 including one or more electronic human behavior simulations. The computing device 104 receives, via the human behavior simulation module 112, an electronic human behavior simulation from the event list 132. The computing device 104 provides, via the human behavior simulation module 112, selected data of the electronic human behavior simulation to memory of the server 104 for recordation (e.g., in memory 136). The computing device 104 executes, via the human behavior simulation module 112, the electronic human behavior simulation on the mobile test device 124. The computing device 104 executes, according to the test script 108, a test case stored in memory of the computing device 104 on the mobile test device 124.
The electronic human behavior simulation can be received by the mobile test device 124 at a random time and/or generated based on a record of a previously run electronic human behavior simulation, with the record stored in memory of the server 116 and including at least one parameter of the previously run electronic human behavior simulation. Executing the electronic human behavior simulation and executing the test case can occur simultaneously or near-simultaneously. The electronic human behavior simulation can simulate many events, such as a human making a phone call, a human receiving a phone call, a human rotating and/or translating a mobile device, a human switching between mobile applications, a human rotating a mobile device such that a display screen of the mobile device changes orientation, the mobile device changing a network connectivity status, or the mobile device receiving a push notification. The method can also include logging, by computing device 104, via a log module in communication with the human behavior simulation module, the selected data of the electronic human behavior simulation in a database of the server 116.
The electronic human behavior simulation can include a physical movement implemented by auxiliary hardware in electronic communication with the mobile device test equipment 120. The mobile device test equipment 120 can include a mobile device test fixture, the mobile device test fixture configured to rotate or translate the mobile test device 124 or to interact with a screen of the mobile test device 124. The mobile device test equipment 120 can include an auxiliary mobile phone used to make a phone call to the mobile test device. The electronic human behavior simulation can include providing a push notification to the mobile test device via a cloud messaging service 140 in electronic communication with the mobile test device 124. The test script can include instructions specifying a time at which to run the human behavior simulation module.
While the test scripts 204 are running individual test cases, the system 200 calls the human behaviors simulation module 208 to generate random events, such as making a phone call, switching applications and/or rotating an application screen. The human behaviors generator 212 will generate these events and log all events into the server 240 via the Log Module 216. For some events, such as switching applications or switching on and/or off of certain networks, the WebDriverAgent/UIAutomator 220 will run these commands (e.g., in iPhones or Android phones). For push notification events, the notification module 224 pushes the message to the mobile test device 244 via the GCM/APNs 232. For events including physical motion of the mobile test device 244, an action is sent via the Internet of Things module 228 to the Mobile Device Test Equipment 236, and then the Mobile Device Test Equipment 236 shifts or rotates the mobile test device 244 physically.
Thus, in this solution, a human behavior generator framework is used to simulate different kinds of human behaviors during mobile application testing. When test scripts are run, human-like behaviors are generated and executed at the same time. This procedure very closely mimics application usage in the real world. In addition, all generated human-like behaviors can be recorded in a database. When a bug is tested out, the records can be retrieved from the database, and the test scripts can be replayed to show how the bug was generated.
It should also be understood that various aspects and embodiments of the technology can be combined in various ways. Based on the teachings of this specification, a person of ordinary skill in the art can readily determine how to combine these various embodiments. In addition, modifications may occur to those skilled in the art upon reading the specification.