Skip to content

Latest commit

 

History

History
379 lines (297 loc) · 20.6 KB

File metadata and controls

379 lines (297 loc) · 20.6 KB

RestAssured-Cucumber-Hybrid-Framework

RestAssured cucumber hybrid framework template

Test execution

  • Test application: https://gorest.co.in/
  • DB setup: https://github.com/abhishekkadavil/localstack

We can execute the test in different ways

  • Locally
    • Through maven
      • mvn clean test to execute test through maven.
      • To execute specific tags from command line we can use mvn test -> -Dcucumber.filter.tags="@All"
    • Through the TestSuitRunner.java class
  • Rerun failed cases: Add rerun:target/failedrerun.txt in the plugin section.

In any point the execution in framework is start from runners.TestSuitRunner.java class. Test that need to execute are defined in the features files present in the Features folder, The set of test cases are identified by appropriate tag name using the tags section in TestRunner class.

Test execution configurations

  • Test API used from https://gorest.co.in/
  • Execution mode(parallel or sequentially): In scenarios method in runners.TestSuitRunner.java class, we can set if the test need to execute parallel or sequentially. Set the parallel flag to true or false.
    • The default test case count can be set from threadCount section in pom file maven-surefire-plugin
  • Report configuration: Test report high level(suit level) configured will be done through @BeforeClass and @AfterClass annotations of TestNG inside TestRunner class. Each scenario wise step will be added to report through utils.handleTestCaseStarted.

RunnerHelper class will be shared among runners.TestRunner and runners.FailedTCRunner class to implement code re-usability.

Test case creation

Test cases are defined in the Features folder in the form of .feature file. We can create test case for any API with some predefined steps. We can create test case in two-way

  • Customisable API test cases - This is long steps and have very much customisable options.
  • API specific test cases - This have shot steps and have to create classes for each API.

Customisable API test cases

  @CreateUserAPI  @CreateUser01 @CreateUserAPIPass @All
  Scenario: Create new user
    #Adding request path
    #Given request have path '<request path>'
    Given request have path '/users'

    #Add query parameters for request
    Given request have below query parameters
      | employeeID | UQS786980d7 |

    #Adding request body from json
    #Given request have request body '<request body path>'
    Given request have request body '/createUser/scenario1/input/requestBody.json'

    #Updating the body with random data like email
    Given request have random email

    #Adding request with headers
    Given request have following headers
      | Content-Type        | application/json |

    #Call post method for the request
    #When I call (POST||DELETE||GET...etc) request
    When I call POST request

    #Retrive from response and strore it in a variable
    #Given retrieve "<json path>" from "<API name>" response and store it in '<unique key where value to be stored>'
    Given retrieve "id" from "CreateUser" response and store it in 'idValue'

    #Put retrieved data in the request path
    #Given request have scenario context '<unique key where value to be stored>' in request path '<req path>'
    Given request have scenario context 'idValue' in request path '/users'

    #Put retrieved data in the request body from value defined in the feature file
    #Given put value "<value to be added in the request>" in path "<JSON path>"
    Given put value "male" in path "gender"

    #Put retrieved data in the request body, from value retrived and stored during the execution
    #Given put scenario context value "<context key which the value stored during the retierval>" in path "<JSON path>"
    Given put scenario context value "female" in path "gender"
    
    # multi part data handling
    # request have following string multi part data
    * request have following string multi part data
      | projectId     | 2                        |
      | awardName     | Builder of the Year 2022 |
    # request have following multi part images
    * request have following multi part images
      | projectAwardPhoto | /Builder/Project/Award/Upload-Award/TC01/input/img1.png |
    # request have following multi part files
    * request have following multi part files
      | projectDocument | /Builder/Project/Documents/Upload-Project-Document/TC01/input/TestPDFfile.pdf |

API specific test cases

Here we have to create specific class for each API eg: createUserStepDef.java

  @E2E  @E2E01 @E2EPass @All_E2E @All
  Scenario: Create user - API specific method 1
    Given start new scenario
    * create user

  @E2E  @E2E02 @E2EPass @All_E2E @All
  Scenario: Create user - API specific method 2
    Given start new scenario
    #Api setup start
    * start create user api specification
    * request have request body '/createUser/scenario1/input/requestBody.json'
    When I call POST request
    Then response code should be '422'

API call chaining - Dynamic request creation

* Add request value from step - Given put value "male" in path "gender"
	* Added API call chaining without context sharing - 
		* create request
		* update specific value in the request
* Add request value from scenario context - @UpdateUserAttribute03(Given put scenario context value "female" in path "gender")
	* Added API call chaining with scenario context sharing - 
		* call a request
		* extract id from the request
		* create another request using the retrieved id
		* Given request have scenario context 'idValue' in request path '/users'
* Add request url parameter from scenario context - @GetUserAPI02 - Given request have scenario context 'idValue' in request path '/users'

Request specification step building logic:

  • Created reqId in ScenarioContext hence sharing is easy between steps. reqId will act as a scenario id key to build request specification map.
  • Generate and put reqId in @Given("start new scenario") using scenarioContext.setReqId(scenarioContext.generateReqId());
  • What will happen during chaining the request?
    • We are using @Given("start new scenario") when ever we start a scenario. When we are chaining two different APIs in a single scenario, We don't have to use @Given("start new scenario") twice i.e we can avoid generating new reqId. Eg: CreateUser02, CreateUser07. If we have duplicate steps in same scenario the latest step will override the value. That is, suppose of we have steps like below
        Given request have path '/users'
        Given request have path '/users/5153'
  • During the first step execution request path should be '/users', When ever the second step execute the request path should be '/users/5153'

Assertions

All the assertion common assertion steps are present in CommonAssertions.java

    #validating response code
    #Then response code should be '<201||400||200 etc>'
    Then response code should be '201'

    #Validating response body with ignoring all extra fields
    And response body should be '/createUser/scenario1/output/responseBody.json' ignoring all extra fields

    #Validating response body with ignoring specified extra fields
    And response body should be '/createUser/scenario3/output/responseBody.json' ignoring specified fields
    | email | id |

    #Validating response body without ignoring all extra fields
    And response body should be '/createUser/scenario2/output/responseBody.json'

    #Validating response body for one field only
    And response should have 'name' as 'Tenali Ramakrishna'

Mock creation

For creating mocks we are using wiremock.

  @MockAPI01 @MockAPIAll
  Scenario: Create new user
    #Mock setup start by adding url to mock
    #Given create (POST||GET||PATCH...etc) mock '<mock name>' to URL '<path>'
    Given create POST mock 'mock' to URL '/users'

    #Add JSON request body to mock
    #Given '<mock name>' external call expects json request body '<request body path>'
    Given 'mock' external call expects json request body '/wiremockFeatures/scenario1/mock/requestBody.json'

    #Add XML request body to mock
    #Given '<mock name>'external call expects xml request body '<request body path.txt>'

    #Add headers to request body
    Given request 'CreateUser' have following headers
      | Content-Type        | application/json |

    #Add JSON response body and status to mock
    #Given '<mock name>' external call with json response body '<response body path>' and status '<status>'
    Given 'mock' external call with json response body '/wiremockFeatures/scenario1/mock/responseBody.json' and status '201'

    #Add XML response body and status to mock
    Given '<mock name>' external call with xml response body '<response body path>' and status '<status>'

    #Add text response body and status to mock
    Given '<mock name>' external call with text response body '<response body path>' and status '<status>'

    #Add text response body and status to mock
    Given '<mock name>' external call with timeout

    #Add add mock to wiremock server
    #Given stub '<mock name>'
    Given stub 'mock'
    #mock setup end

    #We can call the corresponding API with above mock

    #Api setup start
    Given request 'CreateUser' have path '/users'
    Given request 'CreateUser' have request body '/wiremockFeatures/scenario1/input/requestBody.json'
    Given request 'CreateUser' have following headers
      | Content-Type        | application/json |
    When I call POST 'CreateUser' request
    #Assertions
    Then 'CreateUser' should have response code '201'
    And 'CreateUser' should have response body '/wiremockFeatures/scenario1/output/responseBody.json' ignoring all extra fields

    #Mock support: Added url pattern step to mock
    Given create GET mock 'mock' to URL pattern '/abc/def'

    #Mock support: Added response with no body
    Given 'mock' external call status '200'

DB Operations

  @DbOps01 @DbOpsAll
  Scenario: update DB with CSV and validate - SUCCESS - with ignore columns

    #Execute statement
    #Given execute statement "<sql statement>"
    Given execute statement "delete from users"

    #Update DB with CSV
    #Given execute statement "<csv path(users.csv) and table name txt file(table-ordering.txt) path>"
    Given db is updated with "testData/DbOps/scenario1/input/"

Authentication support

Auth token can be passed from maven command line as well as config file. If the maven token parameter is empty then the token will be fetched from config file.

  • Cookie based
    • Passing token through maven : mvn clean install -DCookieToken = authToken=saascsacsac
      Given request have cookie token
  • bearer token
    • Passing token through maven : mvn clean install -Dtoken=68623a1c855aebc18cece732e35d920240db7deaeb49d74581729d57ad940987
      Given request have bearer token in header

DB Assertion

    #Assert DB with CSV
    #Then Expect '<csv path(users.csv) and table name txt file(table-ordering.txt) path>'
    Then Expect 'testData/DbOps/scenario1/output/'

    #Assert DB with CSV by ignoring specified columns
    #Then Expect '<csv path(users.csv) and table name txt file(table-ordering.txt) path>' with ignore columns
    #  | <column name> |
    Then Expect 'testData/DbOps/scenario1/output/' with ignore columns
      | user_id |
      
    #Execute select query to validate data exist in DB
    #Then validate data exist for select query "<sql query>"
    Then validate data exist for select query "select * from users where name = 'abhishek kadavil'"

    #Execute select query to validate data does not exist in DB
    #Then validate data does not exist for select query "<sql query>"
    Then validate data does not exist for select query "select * from users where name = 'abhishek kadavil1'"
    
    #Execute select query from file to validate data exist in DB
    #Then validate data exist for select query from file "<sql file>"
    Then validate data exist for select query from file "/DbOps/scenario9/SelQuery.sql"

    #Execute select query from file to validate data does not exist in DB
    #Then validate data does not exist for select query from file "<sql query>"
    Then validate data does not exist for select query from file "/DbOps/scenario11/SelQuery.sql"
    
    #Execute select query to validate data exist in DB with scenario context
    Then validate data exist for select query '<sql query>' and where condition as a scenario context value '<context value>'
    Then validate data exist for select query 'select * from users where name = ' and where condition as a scenario context value 'name'
    
    #Execute select query to validate data exist in DB with scenario context
    Then validate data not exist for select query '<sql query>' and where condition as a scenario context value '<context value>'
    Then validate data not exist for select query 'select * from users where name = ' and where condition as a scenario context value 'name'
    
    #Execute select query from file to validate data exist in DB
    Then validate data exist for select query from file "<sql query>" and scenario context value '<context value>'
    Then validate data exist for select query from file "/DbOps/scenario18/SelQuery.sql" and scenario context value 'name'

    #Execute select query from file to validate data does not exist in DB
    Then validate data not exist for select query from file "<sql query>" and scenario context value '<context value>'
    Then validate data not exist for select query from file "/DbOps/scenario17/SelQuery.sql" and scenario context value 'name'

Test reporting:

Suit level configuration: Test report high level(suit level) configured will be done through @BeforeClass and @AfterClass annotations of TestNG inside TestRunner class.

Scenario level configuration: Each scenario wise step will be added to report through handleTestCaseStarted from com.utils.TestListener. We only have to create test in extend report in the scenario level configuration that is done using handleTestCaseStarted

Step level configuration: For test step status management are using listener class named com.utils.TestListener which implements cucumber plugin ConcurrentEventListener. Using this plugin we are managing the status of the test cases by monitoring test steps. We are handling three main status FAILED, PASSED, SKIPPED. Since we have all the steps in stepDef package, we added request response and other related details to report through the same classes in the package.

Test logging:

Logging is started in handleTestCaseStarted from com.utils.TestListener. It will start new thread for each test case.

Context sharing

Context sharing is almost similar to session management. Currently, The framework support two level of context sharing scenario level and the test level.

Scenario context - scenario level context sharing

All the features which are common in scenario level like responseContext, reqBodyContext, contextValues, requestBuilder, configuration and wiremock server are done through com.utils.ScenarioContext. i.e. During the execution if we want to share the data between steps or scenarios, we need to use ScenarioContext.

  • ScenarioContext is marked with @ScenarioScoped, so the class will have separate copy of instance for each scenario
  • We are using google-guice for DI

Test context - test level context sharing

if we want to manage data between scenarios, we can use this level of context sharing. even though the framework support test level context sharing, Please avoid making the test case depend on each other. As per cucumber documentation they strongly recommend to avoid dependency between scenarios. This level of context sharing is only used if we need to share common data among all the scenarios.

Current test context contain static properties hence we do not need to worry about object management. We are initialising the data for the test context before the execution start, in RunnerHelper.beforeTestSuit. after the initialisation, we can access the data through-out the execution. Also after the initialisation we can only read the data from the test context, We cannot set the data once the execution start.

Why make it static?

We only have one copy of the class in entire execution, and we do not need to manage state of any property.

Why don't we make the test context read and write throughout the program? Why only make it readonly after initialisation?

Since we are executing our tests in parallel, if we change value of the test context property in middle of the execution, other test cases will be effected by this.

For example consider below scenario:

  • TestA - set test context property USERID as 001
  • TestB - get test context property USERID

if we use test context to support above case, our test case might fail in some execution. Because Our tests execute in parallel and there are chances in TestB execute before TestA. Hence, it will throw null pointer when we are trying to get context property USERID before setting it.

Note:

Currently, cucumber does not support setting priority for test scenario's execution. Since cucumber directly not support priority settings, If we design the test case to depends on each other and execute scenarios in parallel, it will cause inconsistency. Hence, I recommend to avoid adding the data into test context during the execution. it is okay to add data before the scenario start and use the same though-out the execution.

Normal order of execution is determined as follows:

  • Alphabetically by feature file directory
  • Alphabetically by feature file name
  • Order of scenarios within the feature file

Note: the file name sorting is case-sensitive, so uppercase A is executed before a.

Other Features

  • Added google-juice and cucumber-juice for managing the state of class object
  • Added the @ScenarioScoped(the object of a class marked this annotation is only created for one scenario and destroyed after the use)
    • Added functionality of ScenarioContext to accommodate all the common function in a scenario perspective eg: contextValues
  • The output of the test execution like response body, status, DB values etc. can be logged in report for the later use, so didn't create any other mechanism for that. RestAssuredLoggingFilter: Adding rest assured logs to console including all the details of request and response. Configuration: Configurations for the test suits are done through com.utils.ConfigUtil interface which extends org.aeonbits.owner.Config. Suit level configuration are done in ScenarioContext class.

Why and why not

  • RestAsssured
    • Ve: Want to have more control on the framework
    • -Ve: Need good programming knowledge
  • OOPS, used in framework
    • Runnerhelper class
  • Design pattern used
    • KISS
    • DI injection in Scenario context
  • Why not use grass hopper extend report plugin - it's not support cucumber 7, It's not that much flexible as I wanted
  • Why use Google guice instead of pico container or spring
    • google guice - can do DI(object creation) with annotations and have @ScenarioScoped annotation which will make the state management easy
    • pico container - Do not have @ScenarioScoped annotation
    • spring - this is complex
  • Why not Cucumber JUnit - cucumber junit will not allow us to execute scenarios in parallel only feature files in parallel, but TestNG can.
  • Logging - Slf4J
    • For logging there are multiple loggers are available like log4J, JDK logger etc., so once we implement logging in our framework it would be difficult to switch from one library to another. But if we use facade API like Slf4J we can do that easily because we are implementing Slf4J interfaces not logging library like log4J directly. ie we are using Slf4J conjunction with logging libraries(log4J etc.).

Feature need to add