Patent Grant
10073763
This application claims priority to Indian Patent Application No. 201711046829, filed on Dec. 27, 2017, entitled “Touchless Testing Platform,” the entirety of which is hereby incorporated by reference.
This application generally relates to generating and executing automated testing scripts.
Software applications are designed to accommodate a multitude of transactions, where each transaction often requires the performance of a significant number of functions. Testing of software applications may include creating test cases based on requirements and then executing the test cases through, for example, a test script to detect defects. Test cases may be automated using commercial and open source tools to reduce execution time. For example, a regression test suite is a set of test cases, often written in the form of a script, designed to ensure that each of these functions remain accurate and correct following modification of the underlying programing code. Execution of these suites helps to ensure that defects have not been introduced or uncovered in unchanged areas of the codebase as a result of the modification. Moreover, each level of testing (e.g., unit testing, system testing, and acceptance testing) may have its own regression test suite. Providers of these software applications and/or their corresponding services are faced with the problem of having large regression test suites that are executed manually and the need to automate these test suites to function within any one of a number of industry standard automation tools. Once automated, these test cases can be executed repeatedly and frequently, which adds to the amount of testing coverage for the respective application. However, effective and efficient management of such suites of test cases is both time consuming and effort intensive.
Implementations of the present disclosure are generally directed to a touchless automated platform system. More specifically, implementations are directed to a system that creates automated testing scripts based on test cases determined according to requirements documentation, sequences the test cases for execution, and implements defect solutions determined according to the specific application functionalities involved.
In a general implementation, a system includes one or more processors; and a computer-readable storage device coupled to the one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations comprising: receiving a log file that includes log records generated from a code base; processing the log file through a pattern-mining algorithm to determine a usage pattern; generating a graphical representation based on an analysis of the usage pattern; processing the graphical representation through a machine-learning algorithm to select a set of test cases from a plurality of test cases for the code base and to assign a priority value to each of the selected test cases; sequencing the set of test cases based on the priority values; and transmitting the sequenced set of test cases to a test execution engine.
In another general implementation, a computer-implemented method executed by one or more processors includes: receiving a log file that includes log records generated from a code base; processing the log file through a pattern-mining algorithm to determine a usage pattern; generating a graphical representation based on an analysis of the usage pattern; processing the graphical representation through a machine-learning algorithm to select a set of test cases from a plurality of test cases for the code base and to assign a priority value to each of the selected test cases; sequencing the set of test cases based on the priority values; and transmitting the sequenced set of test cases to a test execution engine.
In yet another general implementation, one or more non-transitory computer-readable storage media coupled to one or more processors and having instructions stored thereon which, when executed by the one or more processors, cause the one or more processors to perform operations that include: receiving a log file that includes log records generated from a code base; processing the log file through a pattern-mining algorithm to determine a usage pattern; generating a graphical representation based on an analysis of the usage pattern; processing the graphical representation through a machine-learning algorithm to select a set of test cases from a plurality of test cases for the code base and to assign a priority value to each of the selected test cases; sequencing the set of test cases based on the priority values; and transmitting the sequenced set of test cases to a test execution engine.
In an aspect combinable with any of the general implementations, the operations further include generating a report based on the usage pattern, the failure pattern, the business flow graph, or the anomaly; and transmitting the report to a control center application.
In another aspect combinable with any of the previous aspects, the operations further include analyzing the usage pattern to identify pattern groups; and generating metadata for the identified pattern groups.
In another aspect combinable with any of the previous aspects, the operations further include determining churn data based on an analysis of configuration and coding changes to the code base; processing test results, the churn data, and the code base through an Artificial Intelligence (AI) model, the AI model trained using training data comprising resolutions to reported defects for the code base; and determining a likelihood of failure for at least one of functionalities of the code base, where in the graphical representation is generated further based on the likelihood of failure for the at least one of the functionalities of the code base.
In another aspect combinable with any of the previous aspects, the operations further include demining a failure pattern based on an analysis of testing results prior testing to the code base, wherein the each of the test cases in the set of test cases is further selected based on the failure pattern, and wherein the assigned priority value for each of the test cases in the set of test cases is further determined based on the failure pattern.
In another aspect combinable with any of the previous aspects, the operations further include before receiving the log file, extracting terminologies from requirement documents; classifying the extracted terminologies into categories based on a corpus of known terms; and generating process maps based on the classified terminologies.
In another aspect combinable with any of the previous aspects, the operations further include before selecting a set of test cases, clustering the plurality of test cases for the code base based on contextual similarity.
In another aspect combinable with any of the previous aspects, the operations further include before receiving the log file, receiving a test scenario and a context file selected through a user interface, the context file including an object map comprising objects that correlate to respective components of a display page for the code base, the test scenario describing one of the test cases involving an intended interaction with at least one of the components on the display page; correlating the intended interaction with the at least one component with the corresponding object in the object map; processing the intended interaction and the corresponding object through an AI model, the AI model trained using training data comprising a plurality of processes and respective process steps supported by the components of the display page; determining a script template based on the processing and a selected automation tool; applying, based on the processing, the script template to the intended interaction and the correlated object to generate an automated testing script for the selected automating tool; and assigning the generated automated testing script to the one of the test cases.
In another aspect combinable with any of the previous aspects, the operations further include before receiving the log file, determining a contextual distance for each of the test cases respective to each other; assigning each of the test cases to at least one cluster based on the contextual distances; and assigning a score value to each test cases based on a robustness to detect defects respective to the other test cases in the at least one cluster.
In another aspect combinable with any of the previous aspects, the operations further include determining a resolution for a defect, the defect reported based on a result of an execution a testing script assigned to one of the sequence set of test cases; and implementing the resolution in the code base.
Another aspect combinable with any of the previous aspects, the graphical representation is a heat map.
In another aspect combinable with any of the previous aspects, processing the log file through the pattern-mining algorithm further determines a failure pattern, a business flow graph or an anomaly.
In another aspect combinable with any of the previous aspects, the graphical representation depicts testing priorities of the functionalities of the code base.
In another aspect combinable with any of the previous aspects, the test cases are selected and assigned a priority based on a criterion.
In another aspect combinable with any of the previous aspects, the criterion is a maximization of defect yield per test case executed.
In another aspect combinable with any of the previous aspects, the machine-learning algorithm is a reward-seeking algorithm trained to maximize defect yield per test case.
In another aspect combinable with any of the previous aspects, each of the test scenarios is represented by a process graph that depicts a hierarchy of business process.
In another aspect combinable with any of the previous aspects, the contextual distance for each test case includes a semantic distance for each of the test cases respective to each other, a defect distance for each of the test cases respective to each other, and a code hit distance for each of the test cases respective to each other.
It is appreciated that methods in accordance with the present disclosure can include any combination of the aspects and features described herein. That is, methods in accordance with the present disclosure are not limited to the combinations of aspects and features specifically described herein, but also may include any combination of the aspects and features provided.
The details of one or more implementations of the present disclosure are set forth in the accompanying drawings and the description below. Other features and advantages of the present disclosure will be apparent from the description and drawings, and from the claims.
Application and system testing includes a defined set of activities pertaining to test case design, execution, and defect reporting. With increasing complexity of information technology (IT) architectures, increasing adoption of agile and DevOps, and quicker speed to market, testing practice expectations for overall performance and efficiency of product releases is increased. Thus, testing methodologies to assure the go-live readiness in the shortest possible time of systems and application for release as well as improvements in the quality, stability, and reliability of these methodologies is beneficial. Accordingly, a need exists for widening the mandate for testing as well as employing new strategies and techniques within the creation, maintenance, and execution of regression suites. For example, methodologies may be employed to create test cases based on a set of criteria, such as maximizing testing coverage. The methodologies may meet the criteria by, for example, increasing automation and efficient reporting of discovered defects. Accordingly, a robust testing platform embraces speed and data driven intelligence, which enable users to effectively and agility produce quality results.
In view of the foregoing, implementations of the present disclosure are generally directed to a touchless testing platform system for creating automated testing scripts, sequencing test cases, and implementing defect solutions. The described system employs techniques, such as data mining, artificial intelligence, machine learning, and natural language processing, to build and maintain enterprise testing suites. Beginning with requirements documentation, the described system generates test scenarios, which are used to create test cases once the application or system has been built. Requirements documentation (e.g., business requirements, functional requirements, use cases, user stories, and so forth) captures, for example, information related to the business process(es) that will be supported by the software, intended actions that will be performed through the software, managed data, rule sets, nonfunctional attributes (e.g., response time, accessibility, access and privilege), and so forth. Using this documentation, the described system can be employed to extract key processes and events, feature requirements, non-functional testing, and so forth through, for example, name entity recognition, topic segmentation, part-of-speech tagging, terminology and relationship extraction techniques, semantic clustering, and test case mind maps, which may be generated by employing, for example, semantic graphs. Based on this extracted information, test scenarios are generated. Test scenarios explain, for example, a general process without focusing on implementation. Once the design of the software is available, the test scenarios can be converted into test cases specific to the implementation of the respective business requirement(s).
The described system provides benefits to test designers by quickly identifying the key attributes to be tested as well as their operations and relationships, thus significantly improving the productivity and quality of test cases. For example, natural language processing (NLP) and clustering techniques can be employed to extract the intent from a test case and automatically create code. As an example, logging into an application or program, such as a web application, may require a user name and password or completing a particular form in a web application with a certain set of information. The described system may correlate the intent of use with an object(s) on a selected page or screen of the web application. Once the intent and objects are correlated, the system selects a template(s) corresponding to a designated automation tool to generate the automated testing script for the determined test scenarios. Once a test scenario has been translated to an automated testing script, the test scenarios can be tested frequently and used in a variety of testing processes using the respective automated testing script.
Furthermore, testing automation may realize ‘unattended testing’ through a robust automation framework that provides for better coding practices, such as improved exception handling; portability across environments and applications; and the integration of test data and deployment environments into the automation logic. In some examples, test scripts fail not because of a system failure, but due to environment failure and/or incorrect test data. Accordingly, the described system provides an orchestrator process to provide for the oversight of end-to-end execution of regression suites. For example, the described system may generate alerts on failures that are due to the test environment and/or the particular data used to test the functionality of the particular application or system being tested. These processes can also collect system logs and traces when a failure is encountered.
While it is important for the testing process to maximize coverage of system requirements, the described system includes robust and optimized testing processes that search and discover defects. For example, one objective of testing optimization is to maximize defect yield in relation to the amount of cost (e.g., time) spent on testing. The described system optimizes testing by employing AI and NLP techniques to, for example, remove duplicate test cases and/or test cases exercising the same execution paths. NLP can be described as the ability of a computer program or software to understand human speech as it is spoken or written. NLP makes it possible for an AI program to receive conversational or text input by breaking the respective syntax down to determine the input's meaning and/or determine an appropriate action.
Additionally, the described system optimizes testing by improving the maintainability and re-use of regression suites by identifying re-useable components, such as test step and test cases that can be executed through other test cases.
Furthermore, graphical representations, such as heat maps, may be created and employed for a quantitative analysis and selection of a set of test cases to execute. For example, a heat map may be created by determining system usage and failure patterns that may be extracted from production logs where functionalities that have a higher usage and/or a higher propensity to fail may indicate a condition for increased test priority. Heat maps may also be used to determine functionalities where change is occurring at an increased rate. These types of factors may also indicate a condition for increased test priority. Additionally, the code bases to be tested may be analyzed to set testing priorities based on a determined quality of various modules within the code based.
The described system also provides processes for dynamic test sequencing. These processes select test cases to be executed based on test priority, which may be determined according to the generated graphical representations. In some implementations, the order of test case execution or test sequencing is dynamically determined to maximize various thresholds or criteria, such as defect yield per test case executed. This can be done by, for example, clustering test cases based on machine-learning algorithms that are trained according to previous execution results.
The described system provides processes to increase test case vitality. The vitality of a test case includes its ability to activate fault chains, trigger failures, and/or detect defects. For example, the longer a test case is in use, the more its vitality decreases due to factors, such as inadequate coverage of failure paths or improvements to the functionality that is covered by the test case. The described system also includes processes that conducted test vitality analysis to, for example, remove test cases exhibiting poor vitality, thus improving defect yield per unit test case executed.
The described system also integrates functional and non-functional testing as a testing failure may happen due to functional and/or non-functional root causes. For example, a functional defect may originate from non-functional causes, such as a database timeout leading to an incorrect update on a user interface (UI). In such an example, testers may tag the UI issue as a functional defect. To provide for this integration, the described system includes processes that analyze performance, scalability, stability, recoverability, exception handling, upgrade, and so forth, which are executed throughout the testing cycle and/or in parallel. Furthermore, the described system, uses application monitoring and log mining to build useful insights into the underlying architecture behavior as functional tests are being run. For example, thresholds for known problem patterns and monitor alerts may be set and machine learning employed to determine typical system behavior as well as search for anomalies.
The described system also includes processes that continuously build data driven insights throughout the tested systems lifecycle. These insights allow for the implementation of both corrective and prevention actions to, for example, achieve projects goals. Insights may include usage patterns, such as how a system is being used, UI pathways being exercised most often, and the control and data flow pathways that are most prevalent; failure prediction and modeling, such as the system components most likely to fail, the most likely pathways that activate internal fault chains and trigger failures, typical system and run time behavior mining and anomaly detection, and anomalies that are prognostic of failures; churn patterns, such as the modules that are undergoing that most churn, from where/who a change(s) is originating, the modules undergoing burst changes; code and architectural quality, such as whether churn is affecting code and/or architectural quality; defect analysis, such as the likely root causes of defects; and fault chain relationships, such as defects causally relationships to one another, root defects versus symptoms, and duplicates.
The system 100 provides accesses to the users 112 through the control center 110, which may be accessed over a network (not shown). In the depicted system 100, a user 112 may analyze requirements for an application or system and then design, develop, and test software by applying the theories and principles of computer science and mathematical analysis. The control center 110 includes a series of screens, which may be received and viewed on a user device using, for example, a web browser program. Pages may be generated on a server or virtual server and transmitted via a network. The control center 110 allows the user(s) 112 to provide testing scenarios (e.g., test cases) and control commands to the various modules, which are described in detail below, included in the testing platform module 120 and to the data and artifact repository 130. The control center 110 also allows the user(s) 112 to analysis and view testing results.
The data and artifact repository 130 may be a relational database, such as Oracle or DB2, implemented on an enterprise database server or virtual server. The data and artifact repository 130 may store testing data, such as system requirements 131, which include requirements for the application or system that is being tested; defect and past resolution data 132, which includes defects reported and the respective resolution(s); configuration and control data 133, which includes configurations for the tested application or systems; production data 134, which includes data from the production or other deployment environments for the tested system or application; test cases 135, which include test cases for the function features of the tested application or system; test results 136, which include the results of the regression suites and various test scripts; application logs 137, which include logs from the various deployment environments for the tested application or system; trace data 138, which includes information about the tested application or system's execution; and code quality analysis data 139, which includes results of quality analysis for the tested application or system. These various types of testing data 131-139 are provided as examples. Other types of testing data may be stored in the data and artifact repository 130 and used by the control center 110 and/or the testing platform module 120.
The testing platform module 120 includes test scenario and process map extractor engine 121, automation accelerator engine 122, test suite analyzer engine 123, log analyzer and pattern miner engine 124, test priority graphical representation generator engine 125, dynamic test case selector and sequencer engine 126, and defect analyzer engine 127. The testing platform module 120 achieves a seamless automated workflow of a testing lifecycle guided by data driven intelligence. This data driven intelligence is achieved by AI and machine-learning techniques that accelerate the automation of the activities and the decision making at various stages of testing. The testing platform module 120 integrates these modules as well as third-party and open source tools.
The test scenario and process map extractor engine 121 scans the requirements document and creates the high level test scenarios and process maps. The automation accelerator engine 122 analyzes manual test cases, extracts the intent, and converts the intent into executable automated scripts. The test suite analyzer engine 123 analyzes test suite and groups contextually similar test cases into clusters based on contextual distance. The log analyzer and pattern miner engine 124 ingests the log files from deployment environments, such as production or test environment and extracts various insights, such as usage and/or failure patterns, typical system behaviors, and/or anomalies (which in turn can be used as early warning of potential failures). The test priority graphical representation generator engine 125 generates graphical representations, such as heat maps, that depict the test priority of various functionalities. The graphical representation may be generated based on usage patterns, failure patterns, module churn, code quality analysis, or a combination thereof. The dynamic test case selector and sequencer engine 126 selects a next set of test cases to be executed based on a configurable criterion. The defect analyzer engine 127 analyzes defect data and provides data driven insights. Each of these engines are described in greater detail below with regard to
Terminology extractor module 210 extracts the various terminologies from the requirements document by using techniques, such as terminology extraction. For example, the terminology extractor module 210 may pick up key terminologies pertaining to business processes, operations, and/or data and control. This information is fed into, for example, one of the three classifier modules: business process classifier module 220, operations classifier module 222, and data set classifier module 224. Each of the three classifiers may classify the received terms into, for example, business processes, operations and actions, or data sets by using techniques, such as topic segmentation. The business process classifier module 220 classifies terminology pertaining to, for example, a business process described in the requirements document. The operations classifier module 222 classifies terminology pertaining to, for example, business operations applicable to business processes classified by the business process classifier module 220. The data set classifier module 224 classifies data and control information requirements to perform the operations classified by the operations classifier module 222. Each of the three classifier modules may also use entity term corpus 151, which is stored in data and artifact repository 130, as per testing requirements. The entity term corpus 151 includes business process terms, operations terms, technology terms, and data set terms that are domain specific and/or related to the scope of the respective requirements documentation. Once classified, the data is fed into the three builder modules. Three classifier modules are shown and described; however, other classifier modules may be employed in the test scenario and process map extractor engine 121.
The process map builder module 230 builds a process map (e.g., graph) using techniques, such as relationship extraction and/or semantic graphs. The process map may depict, for example, a hierarchy of business processes. The test scenario map builder module 232 builds a testing scenario using techniques, such as knowledge representation. The testing scenario can be depicted using a knowledge graph that may depict, for example, test scenarios, operations applicable to a respective process, and data sets required for such operations. The non-functional attribute builder module 234 identifies the requirements that need non-functional attributes (e.g., quality attributes) and extracts those attributes using techniques, such as relationship extraction. This information is passed to the users 122 by way of the control center 110.
In the depicted example 200, terminology extractor module 210 extracts terms 262 from requirements documentation. Operations classifier module 270 parses these extracted terms 262 and classifies them as/according to process terms 272, operations terms 274, and data set terms 276. Semantic graph builder module 280 uses the classified terms to construct a semantic graph. Process map builder module 282 uses the semantic graphs to construct process maps 284. Test scenario map builder 290 uses the generated semantic graphs and process maps to generate test scenario maps 292.
Terminology extractor module 260 is substantially similar to terminology extractor module 210 from
Test scenario element classifier module 270 classifies the extracted terms 262 into process terms 272, operations terms 274, and data set terms 276 by employing techniques, such as, topic segmentation and so forth. Process terms 272 include terminology pertaining to, for example, a business process identified in the requirements documentation. Operations terms 274 include, for example, business operations applicable to the identified business processes. Data set terms 276 include information regarding requirements to perform the identified operations, such as what data a particular form requires or what type of data is needed for a parsing script to execute. Test scenario element classifier module 270 may employ a corpus of existing known terms to assist in identifying terms and classifying them accordingly.
Semantic graph builder module 280 processes classified terms to construct standardized, grammatically accurate, and non-ambiguous statements. For example, a requirement document may include language stating “The Foreign currency accounting modules should be able to produce general forex voucher which can support following 3 types of transactions: debit advice voucher, forex receipt voucher, and currency swapping voucher.” A semantic graph for this line of requirement built for this information may include a node for “Foreign currency accounting” that is linked to a “general forex voucher” node, which is linked to nodes for “debit advice voucher,” “forex receipt voucher,” and “currency swapping voucher.”
Process map builder module 282 employs the semantics graphs generated by semantic graph builder module 280 to generate process flow maps 284. A process flow map includes information as to how various processes are related to each other. Such as, how processes are hierarchically related. Example process flows include business processes that the respective application integrates with or manages. Other examples process flows include data flow processes, functionality, workflows, blockchains, and so forth. Each generated process flow map 284 may also include details of regarding operations and flow and control of data related to the respective process. The generated process flow maps may be stored in process map repository 286. In some implementations, user 112 reviews the generated process maps to update the information and/or merge them into the other data contained in the process map repository 286.
Test scenario map builder 290 uses the generated semantics graphs and process maps to generate test scenarios maps 292 for the respective requirements documentation. The semantics graphs and process maps include, based on the requirements documentation, processes and functionality that may be tested for an application, valid and invalid operations for each functionality, expected outputs, and the relationships, such as a hierarchically relationship, between the various processes.
The generated test scenario graphs may include, nested test scenarios. These scenarios (or groups of scenarios) can be reused across multiple requirements and applications. For example, login into the application” is a scenario that can be used in a “search for product” scenario and the “search for product” scenario may be part of a “compare products” scenario. The test scenario map builder 290 modularizes the requirements statements from a semantic graph(s) and converts them into complete statements through the use of, for example, NLP. For example, requirements statements may be parsed to extract Intent, Objects, Actions, and/or expected results. The test data elements are also identified where available. Combinations of these are then used by the test scenario map builder 290 to build the complete statements. Test scenario map builder 290 then classifies functional and non-functional test scenarios. The functional scenarios may be used by the automation accelerator engine 122 to generate automation test scripts. The non-functional scenarios may be employed by a testing team(s) to generate test cases specific to their respective areas, such as performance, security, and architectural testing. In some implementations, the test scenario map builder 290 includes a non-functional attribute builder that scans the requirements document and extract requirements that are likely to have performance and scalability requirements. These non-functional attributes, such as response time, concurrent user limits, wait time, page, component load time, and so forth, are identified along with the specified values. These attributes are employed to build, for example, performance and load testing models. The generated test scenarios stored in data and artifact repository 130 are used by the automation accelerator engine 122, as described below.
For example, the automation accelerator engine 122 generates an automated testing script(s) from a provided test scenario and context file. The automation accelerator engine 122 extracts the intended interaction (intent) and relevant testing data from each test scenario through the employment of, for example, natural language processing (NLP) techniques. The intent is correlated to an appropriate test object(s) in the provided context file. For example, if the test scenario recites “Click on the Submit button,” the automation accelerator engine 122 parses the natural language and derives the context as “submit button,” which it then maps to the submit button object from the object map of the submitted context file. A template for the selected automation tool is applied to the extracted intent and data along with the correlated object(s) to generate the resulting automated testing script.
At a high level, the NLP Engine 310 employs NLP to parse and extract the intent from manual test cases. The object correlator module 330 creates the logical objects in accordance with the extracted intent, which it maps to objects in the context file. The application object scanner module 320 scans the tested application or system and identifies the objects within, for example, the UI pages. The script generator module 340 generates the test scripts bases on the input from object correlator module 330.
For example, test cases (e.g., test scenarios) may be described in a free flow language form, without any well-defined format, and in the natural or agreed upon language of the parties, such as English. Thus, it is difficult for a computer program to translate this information into an automated testing script. Moreover, development projects are increasingly using test scenarios written in a behavior-driven development (BDD) style using a formatted language, such as Gherkin. The formatted language allows for instructions to be written as plain text in a traditional written language, such as English or one over 60 other languages, with some additional structure. These formatted languages are designed to be easy to learn by non-programmers, yet structured enough to allow concise descriptions of examples to illustrate business rules in most real-world domains. In this way, the formatted languages can be used to capture requirements and define the test scenarios.
By way of example, a test scenario or list of scenarios may be included in what is called a feature file, where a formatted language, such as Gherkin, is used to write the scenarios in a human readable way. Such a feature file may be used in the generation of an automated testing script for an automation tool. Example automation tools include Unified Functional Testing (UFT), Tricentis Tosca™, Worksoft Certify™, and Selenium™. The testing automation tool provides a framework that can be used to provide support software structures, such as step definitions, for each of the test scenarios. Step definitions act as skeleton placeholders where automation code blocks may be implemented. For example, each step in a given scenario may map to a step definition. The automation code block is implemented for each step definition and executed when the scenario is run by the testing framework. The automation code block may be written in a variety of programming language, such as Ruby, C++, Java, Scala, Python, and so forth, selected based on system requirements. Once generated, the step definitions and respective code blocks, may be referred to as an automated testing script. The testing automation tool provides an execution environment for these generated scripts, which may be run for acceptance and/or regression testing.
The automated testing scripts may be implemented manually by a developer or generated automatically. One of the difficulties with automatic code generation of an automated testing script from a test scenario is that the test scenario may be written in many different ways because of the flexible nature of the formatted language (e.g., Gherkin). For example, each tester may describe the functionality or function of a test in his or her own way. For instance, “click on the submit button,” “push the submit button,” or “hit the submit button,” all of which mean the same thing. Another issue is that the code blocks in an automated testing script may be repetitive, hence, the automation code may become bloated. Additionally, the free form English (or any other language), which may be used in the bulk of the file, lacks structure. For example, “Login to the application” and “Enter the user name, enter the password, and click sign in button” both denote the same function; one is a single sentence, but the other is three sentences. However, the same automation code block (script) should be generated in each of these examples.
The automation accelerator engine 122 parses a provided test scenario base on natural language processing techniques to determine the scenario's intent for the code base that is being testing along with any data relevant to the scenario. As an example, a scenario's intent may be to login to a given application or program, such as a web application, with a username and password or to complete a particular form in the web application with a certain set of information. In the proceeding example the username and password and the set of information are the data relevant to the scenario. The system correlates the intent with an object(s) on a selected page or screen of the web application. Once the intent and objects are correlated, the system selects a template(s) corresponding to a designated automation tool to generate the automated testing script for the provided scenario. Once the test scenario has been translated to an automated testing script, the test scenarios can be tested frequently and used in a variety of testing processes through the use of the respective automated testing script. Additionally, the automated testing script may be more precise than the manual versions and may also be used in the generation of reports regarding respective application and/or the results of the testing.
In the depicted example, the NLP engine 310 receives a test scenario(s) from the data and artifact repository 130, and optionally, a context file from the control center 110. The received test scenarios may be generated by the test scenario and process map extractor 121 engine as described above. The test scenarios may also include existing test cases, feature files, API definition files, such as Web Services Description Language (WSDL), Web Application Description Language (WADL), Swagger, and so forth. The NLP engine 310 receives the input and parses the test scenarios. NLP can be described as the ability of a computer program or software to understand human speech as it is spoken or written. The NLP engine 310 may employ an NLP application program interface (API), such as Apache OpenNLP™.
As an example, the parser module 312 reads a line or set of lines from the received test scenario, which may be included in a feature file. The parser module 312 determines the various objects in the sentence(s) in the feature file. The objects are used to provide a context for the instructions in the test scenario. Based on this information, the intent extractor module 314 determines the intent of the scenario for each of the identified objects. For example, the scenario might read “to login, to enter a username, to enter a password, to click on submit, enter navigation.” The intent extractor module 314 extracts the various intents for this scenario (e.g., “click,” “login,” and “enter”). Once the intent has been identified, it may be correlated to the object(s) in the context file by the object correlator module 330 based the context file received from the application object scanner module 320 and selected through the control center 110 by users 112.
Custom actions can also be trained into the intent parser module 312 for complex descriptions. The intent extractor module 314 also identifies, for each determined intent, any associated data, for example, a username and password. Techniques such as text parsing, tokenizer, and name finder can be used to identify the mention of custom data within the test scenario as data specific words may be mentioned in any part of a sentence.
The object correlator module 330 takes the identified intent and any associated data and correlates the information to objects within the context file. The object correlator module 330 first generates a name for the intent based on the NLP processing. The name is based on the context provided in the test scenario, and is referred to as a logical name. The object correlator module 330 searches the object map in the context file to determine the best matching object. For example, a test scenario regarding a web application selling various products may have a movie store section. A test scenario, for example, a login to the movie store, may refer to the section of the site as “DVD movie store.” The intent extractor module 314 may determine the intent for the scenario as “enter the DVD movie store,” where “DVD movie store” is the logical name assigned to the object. The object correlator module 330 takes the assigned logical name and searches the object map from the context file for an object that has an identifier that matches the logical name or is the closest match to the object's name. This match is determined based on the determined intent, NLP of the objects, and any corresponding data in the context file (e.g., identifier and/or associated parameters). The object correlator module 330 correlates the intent (e.g., login, enter, click) to the matched object from the context file. For the movie store example, the object correlator module 330 may return an object corresponding to a button or link that takes the user to the DVD movie store section of the site, which is then correlated with the intent or action of the scenario (e.g., click on the movie store link).
The script generator module 340 generates an automated testing script based on the determined intent and the associated data from the provided scenario and the correlated object(s) from the provided context file. For example, the determined intent could be at an elementary level (click, enter) or a business process level (login, fill out a form). A login may require a series of steps, such as 1) enter user name, 2) enter password, and 3) click sign on. Another example may be to create a purchase order, which could entail filling out a form with multiple steps.
To generate the file, the script generator module 340 may employ an AI model trained through a series of machine-learning techniques applied to an algorithm using these elementary and business level steps. Machine learning automates model building and allows for the discovery of insights without being explicit programming. Using automated and iterative algorithms, models may be trained to find high-order interactions and patterns within data. The AI model may be trained for a particular application or program to apply action(s) required to complete various tasks or processes within the application. For example, an AI model may be trained to understand what a purchase order means, what login means, and how to perform each within the particular application or program. In some implementations, the provided context file is used to determine the appropriate AI model to employ to build the resulting automated testing script.
The script generator module 340 selects a script template(s) from the data and artifact repository 130. In some implementations, the script templates are a standardized form for automation, which may be employed in keyword driven, data driven and/or hybrid automation frameworks. In some implementations, the script templates are standard or proprietary templates defined by an automation tool, such as Tricentis Tosca™. The script template is selected based on the automation tool for which the automated testing script is being generated (e.g., UFT, Tricentis Tosca™, Worksoft Certify™, or Selenium™). Based on the AI model, the script generator module 340, determines the action(s) to perform the determined intent to the correlated objects in the respective page of the UI being tested. The script generator module 340 generates the automated script by applying the selected template to the determined actions for the intent and correlated objects, the data read from the provided test scenario, and the step definitions from the test scenario (e.g., the feature file). The data may also be read from a configuration or properties file. This data may be used as the default data unless specific information is provided in the test scenario. In some instances, the actions may not require any additional data, such as, for example, when simply following a link or clicking a button of a particular page within the UI.
The contextual distance calculator module 410 calculates the contextual distance between a test case and other test case. The contextual distance calculator module 410 includes text and semantic distance calculator module 412, which determined the similarities of test cases based on, for example, textual content and semantics; defect distance calculator module 414, which determines the similarities of test cases based on, for example, the defects each respective test cases is detecting; and the code hit distance module 416, which determines the similarities of test cases based on, for example, the portions of code being executed by the test cases. The aggregate similarity cluster builder module 420 receives the calculated contextual distance information regarding the test cases and constructs similarity clusters. The vitality analyzer module 430 then takes the clustered data to analyze each test case and its robustness to detect defects based on, for example, past history. For example, each test case may be assigned a score value based on its robustness to detect defects respective to the other test cases in an assigned cluster. This information is passed to the users 122 by way of the control center 110.
The log qualifier module 510 qualifies the collected logs from various sources, such as application logs, server logs and UI logs. The algorithm selector module 520 receives the qualified logs from log qualifier module 510 and selects a pattern-mining algorithm based on the initial qualification. The pattern analyzer module 530 receives the selected pattern and qualified logs to analyzes patterns present in the logs. The metadata creator module 540 receives the information regarding patterns from the pattern analyzer module 530 and creates metadata for each of the pattern groups. The data extractor module 560 receives the information from the prior modules and extracts the various data of interest from the identified patterns. The report generator module 560 receives the determined information from the other modules and generates reports on, for example, usage patterns, failure patterns, business flow graphs, and anomalies. This information is passed to the test priority graphical representation generator engine 125.
The module churn analyzer module 610 continuously analyzes configuration and control as well as module churn data 133. The test results analyzer module 620 continuously analyzes the results of execution runs 135 and extracts failure patterns. The code quality analyzer module 630 analyzes static code quality based on preset rules. The static code can be received from code repository 152, which can be stored in data and artifact repository 130. The failure prediction module 640 receives the extracted and analyzed information from modules 610-630 and employs machine-learning algorithms to predict which functionalities in the tested application or system are likely to fail. The graphical representation generator module 650 generates graphical representation depicting relative test priorities of functionalities and associated test cases based on the prediction from module 640 and the received information from modules 610-640. The generated graphical representations are passed to the dynamic test case selector and sequencer engine 126.
The test case selector module 710 selects test cases to be executed from test priority graphical representations received from the test priority graphical representation generator engine 125. The test case cluster builder module 720 clusters the selected test cases based on contextual similarity. The test results analyzer 730 continuously analyzes the results of execution runs and extracts failure patterns 136. Test case sequencer module 740 receives the selected and clustered test cases as well as the execution results and uses, for example, reward seeking machine-learning algorithms to set execution sequence of the clustered test case based on a configurable criterion, such as to maximize defect yield per test case. This sequence of test cases is passed to the test execution engine 140.
The test execution engine 140 includes an auto-script server that runs the automation tools/platforms, such as Unified Functional Testing (UFT), Tricentis Tosca™, Worksoft Certify™, or Selenium™, employed within the touchless testing platform system 100. Such automation tools are used to provide stakeholders with information about the quality of the application or system being tested. The test execution engine 140 may be deployed on a server or virtual server.
The similarity cluster builder module 810 clusters defects 132 based on similarity analytics. The defect analyzer module 820 classifies the clustered defects based on an AI module trained through machine learning with, for example, past resolution data. This determined recommendation and results of executed resolutions are passed users 112 by way of the control center 120. In some implementations, the dynamic defect analyzer engine 127 may determine, through the defect analyzer module 820, a resolution for a defect reported based on a result of an execution of a testing script assigned to one of the sequence set of test cases and implement the resolution in the code base for the application or system to be tested.
The control center 120 may provide this information to the users 112 through a reporting engine which provides analytics and access to the reporting features. The execution engine 160 may persist results from the execution of the generated automated testing scripts in a reporting database (not shown). The reporting engine may generate reports from the information stored in the reporting database, which can be reviewed by users 112. These reports provide details on the performance of the system during the execution of the automated testing scripts and may include processing time, response time, and any warning messages displayed as well as the information generated by the various engines 131-137 of the touchless testing platform module 120. Statistical reports may also be generated that summarize the health of the system as well as the any significant data points selected.
The computing device 1000 includes a processor 1002, a memory 1004, a storage device 1006, a high-speed interface 1008 connecting to the memory 1004 and multiple high-speed expansion ports 1010, and a low-speed interface 1012 connecting to a low-speed expansion port 1014 and the storage device 1006. Each of the processor 1002, the memory 1004, the storage device 1006, the high-speed interface 1008, the high-speed expansion ports 1010, and the low-speed interface 1012, are interconnected using various buses, and may be mounted on a common motherboard or in other manners as appropriate. The processor 1002 can process instructions for execution within the computing device 1000, including instructions stored in the memory 1004 or on the storage device 1006 to display graphical information for a GUI on an external input/output device, such as a display 1016 coupled to the high-speed interface 1008. In other implementations, multiple processors and/or multiple buses may be used, as appropriate, along with multiple memories and types of memory. Also, multiple computing devices may be connected, with each device providing portions of the necessary operations (e.g., as a server bank, a group of blade servers, or a multi-processor system).
The memory 1004 stores information within the computing device 1000. In some implementations, the memory 1004 is a volatile memory unit or units. In some implementations, the memory 1004 is a non-volatile memory unit or units. The memory 1004 may also be another form of computer-readable medium, such as a magnetic or optical disk.
The storage device 1006 is capable of providing mass storage for the computing device 1000. In some implementations, the storage device 1006 may be or contain a computer-readable medium, such as a floppy disk device, a hard disk device, an optical disk device, or a tape device, a flash memory or other similar solid state memory device, or an array of devices, including devices in a storage area network or other configurations. Instructions can be stored in an information carrier. The instructions, when executed by one or more processing devices (for example, processor 1002), perform one or more methods, such as those described above. The instructions can also be stored by one or more storage devices such as computer- or machine-readable mediums (for example, the memory 1004, the storage device 1006, or memory on the processor 1002).
The high-speed interface 1008 manages bandwidth-intensive operations for the computing device 1000, while the low-speed interface 1012 manages lower bandwidth-intensive operations. Such allocation of functions is an example only. In some implementations, the high-speed interface 1008 is coupled to the memory 1004, the display 1016 (e.g., through a graphics processor or accelerator), and to the high-speed expansion ports 1010, which may accept various expansion cards. In the implementation, the low-speed interface 1012 is coupled to the storage device 1006 and the low-speed expansion port 1014. The low-speed expansion port 1014, which may include various communication ports (e.g., USB, Bluetooth, Ethernet, wireless Ethernet) may be coupled to one or more input/output devices. Such input/output devices may include a scanner 1030, a printing device 1034, or a keyboard or mouse 1036. The input/output devices may also by coupled to the low-speed expansion port 1014 through a network adapter. Such network input/output devices may include, for example, a switch or router 1032.
The computing device 1000 may be implemented in a number of different forms, as shown in the
The mobile computing device 1050 includes a processor 1052, a memory 1064, an input/output device such as a display 1054, a communication interface 1066, and a transceiver 1068, among other components. The mobile computing device 650 may also be provided with a storage device, such as a micro-drive or other device, to provide additional storage. Each of the processor 1052, the memory 1064, the display 1054, the communication interface 1066, and the transceiver 1068, are interconnected using various buses, and several of the components may be mounted on a common motherboard or in other manners as appropriate.
The processor 1052 can execute instructions within the mobile computing device 1050, including instructions stored in the memory 1064. The processor 1052 may be implemented as a chipset of chips that include separate and multiple analog and digital processors. For example, the processor 1052 may be a Complex Instruction Set Computers (CISC) processor, a Reduced Instruction Set Computer (RISC) processor, or a Minimal Instruction Set Computer (MISC) processor. The processor 1052 may provide, for example, for coordination of the other components of the mobile computing device 1050, such as control of UIs, applications run by the mobile computing device 1050, and wireless communication by the mobile computing device 1050.
The processor 1052 may communicate with a user through a control interface 1058 and a display interface 1056 coupled to the display 1054. The display 1054 may be, for example, a Thin-Film-Transistor Liquid Crystal Display (TFT) display or an Organic Light Emitting Diode (OLED) display, or other appropriate display technology. The display interface 1056 may comprise appropriate circuitry for driving the display 1054 to present graphical and other information to a user. The control interface 1058 may receive commands from a user and convert them for submission to the processor 1052. In addition, an external interface 1062 may provide communication with the processor 1052, so as to enable near area communication of the mobile computing device 1050 with other devices. The external interface 1062 may provide, for example, for wired communication in some implementations, or for wireless communication in other implementations, and multiple interfaces may also be used.
The memory 1064 stores information within the mobile computing device 1050. The memory 1064 can be implemented as one or more of a computer-readable medium or media, a volatile memory unit or units, or a non-volatile memory unit or units. An expansion memory 1074 may also be provided and connected to the mobile computing device 1050 through an expansion interface 1072, which may include, for example, a Single in Line Memory Module (SIMM) card interface. The expansion memory 1074 may provide extra storage space for the mobile computing device 1050, or may also store applications or other information for the mobile computing device 1050. Specifically, the expansion memory 1074 may include instructions to carry out or supplement the processes described above, and may include secure information also. Thus, for example, the expansion memory 1074 may be provided as a security module for the mobile computing device 1050, and may be programmed with instructions that permit secure use of the mobile computing device 1050. In addition, secure applications may be provided via the SIMM cards, along with additional information, such as placing identifying information on the SIMM card in a non-hackable manner.
The memory may include, for example, flash memory and/or non-volatile random access memory (NVRAM), as discussed below. In some implementations, instructions are stored in an information carrier. The instructions, when executed by one or more processing devices (for example, processor 1052), perform one or more methods, such as those described above. The instructions can also be stored by one or more storage devices, such as one or more computer- or machine-readable mediums (for example, the memory 1064, the expansion memory 1074, or memory on the processor 1052). In some implementations, the instructions can be received in a propagated signal, for example, over the transceiver 1068 or the external interface 1062.
The mobile computing device 1050 may communicate wirelessly through the communication interface 1066, which may include digital signal processing circuitry where necessary. The communication interface 1066 may provide for communications under various modes or protocols, such as Global System for Mobile communications (GSM) voice calls, Short Message Service (SMS), Enhanced Messaging Service (EMS), or Multimedia Messaging Service (MMS) messaging, code division multiple access (CDMA), time division multiple access (TDMA), Personal Digital Cellular (PDC), Wideband Code Division Multiple Access (WCDMA), CDMA2000, or General Packet Radio Service (GPRS), among others. Such communication may occur, for example, through the transceiver 1068 using a radio-frequency. In addition, short-range communication may occur, such as using a Bluetooth, Wi-Fi, or other such transceivers. In addition, a Global Positioning System (GPS) receiver module 1070 may provide additional navigation- and location-related wireless data to the mobile computing device 1050, which may be used as appropriate by applications running on the mobile computing device 1050.
The mobile computing device 1050 may also communicate audibly using an audio codec 1060, which may receive spoken information from a user and convert it to usable digital information. The audio codec 1060 may likewise generate audible sound for a user, such as through a speaker, e.g., in a handset of the mobile computing device 1050. Such sound may include sound from voice telephone calls, may include recorded sound (e.g., voice messages, music files, etc.) and may also include sound generated by applications operating on the mobile computing device 1050.
The mobile computing device 1050 may be implemented in a number of different forms, as shown in the figure. For example, it may be implemented as a cellular telephone 1080. It may also be implemented as part of a smart-phone, personal digital assistant, or other similar mobile device.
Various implementations of the systems and techniques described here can be realized in digital electronic circuitry, integrated circuitry, specially designed application specific integrated circuits (ASICs), computer hardware, firmware, software, and/or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be for a special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.
These computer programs (also known as programs, software, software applications or code) include machine instructions for a programmable processor, and can be implemented in a high-level procedural and/or object-oriented programming language, and/or in assembly/machine language. As used herein, the terms machine-readable medium and computer-readable medium refer to any computer program product, apparatus and/or device (e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs)) used to provide machine instructions and/or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term machine-readable signal refers to any signal used to provide machine instructions and/or data to a programmable processor.
To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a cathode ray tube (CRT) or liquid crystal display (LCD) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.
The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a client computer having a graphical UI or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.
The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.
Although a few implementations have been described in detail above, other modifications are possible. For example, while a client application is described as accessing the delegate(s), in other implementations the delegate(s) may be employed by other applications implemented by one or more processors, such as an application executing on one or more servers. In addition, the logic flows depicted in the figures do not require the particular order shown, or sequential order, to achieve desirable results. In addition, other actions may be provided, or actions may be eliminated, from the described flows, and other components may be added to, or removed from, the described systems. Accordingly, other implementations are within the scope of the following claims.
| Number | Date | Country | Kind |
|---|---|---|---|
| 201711046829 | Dec 2017 | IN | national |
| Number | Name | Date | Kind |
|---|---|---|---|
| 9032360 | Cohen | May 2015 | B1 |
| 20060156287 | Vikram | Jul 2006 | A1 |
| 20120197595 | Gouthaman | Aug 2012 | A1 |
| 20130074051 | Freeman | Mar 2013 | A1 |
| 20130097586 | Chandra | Apr 2013 | A1 |
| 20150081614 | Bechet | Mar 2015 | A1 |
| 20150378873 | Prasad | Dec 2015 | A1 |
| 20160011960 | Krishna | Jan 2016 | A1 |
| 20160077956 | Bhattacharya | Mar 2016 | A1 |
| 20170344467 | Yadav | Nov 2017 | A1 |
| Entry |
|---|
| Chao Chen, et. al, “Usage-pattern Based Statistical Web Testing and Reliability Measurement”, ScienceDirect Procedia Computer Science 21 (2013) 140-147. |
| U.S. Appl. No. 15/867,293, filed Jan. 10, 2018, Girish Kulkarni. |
