<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Anandhi K on Medium]]></title>
        <description><![CDATA[Stories by Anandhi K on Medium]]></description>
        <link>https://medium.com/@anandhik?source=rss-bc7bd0890b9f------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*2CBzFHBgIGJ_Xs3U6VNdOQ.jpeg</url>
            <title>Stories by Anandhi K on Medium</title>
            <link>https://medium.com/@anandhik?source=rss-bc7bd0890b9f------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Tue, 17 Nov 2026 01:47:11 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@anandhik/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Integrate Cypress, LambdaTest and GitLab Pipeline]]></title>
            <link>https://anandhik.medium.com/integrate-cypress-lambdatest-and-gitlab-pipeline-77d54897d77b?source=rss-bc7bd0890b9f------2</link>
            <guid isPermaLink="false">https://medium.com/p/77d54897d77b</guid>
            <category><![CDATA[cypressgitlab]]></category>
            <category><![CDATA[gitlabpipeline]]></category>
            <category><![CDATA[cypresslambdatest]]></category>
            <category><![CDATA[gitlabcypresslambdatest]]></category>
            <category><![CDATA[gitlab-cicd]]></category>
            <dc:creator><![CDATA[Anandhi K]]></dc:creator>
            <pubDate>Mon, 01 Jun 2026 09:49:57 GMT</pubDate>
            <atom:updated>2026-06-01T09:49:57.717Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/576/0*YffiQjjegBENhgVY" /></figure><p>This article is about running Cypress tests in LambdaTest environment using GitLab pipeline.</p><p>Going to follow below steps to achieve this.</p><ol><li>Install cypress lambdatest</li><li>Configure lambdatest credentials in local file</li><li>Run locally</li><li>Push project into GitLab Repo</li><li>Configure CI/CD pipeline variables</li><li>Create .gitlab.yml file</li><li>Run pipeline job</li></ol><p>Prerequisites :</p><p>Create a project and install Cypress</p><p>Create a test to perform some scenario</p><p>Verify whether it is successfully running.</p><p>Step 1: Install cypress lambdatest</p><p>&gt; npm install -g lambdatest-cypress-cli</p><p>To generate lambdatest config file run,</p><p>&gt; lambdatest-cypress init — cv=14 (depends on your cypress version)</p><p>This will automatiocally generate Lambdatest.config.json, lambdatest_run.json, and base_reporter_config.json files.</p><p>Step 2 : Configure lambdatest credentials in local file</p><p>Login into Lambdatest, generate access-key and configure lambdatest.config.json file as below,</p><pre>{<br>   &quot;lambdatest_auth&quot;: {<br>      &quot;username&quot;: &quot;XXXXX&quot;,<br>      &quot;access_key&quot;: &quot;xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx&quot;<br>   },<br>   &quot;browsers&quot;: [<br>      {<br>         &quot;browser&quot;: &quot;Chrome&quot;,<br>         &quot;platform&quot;: &quot;Windows 10&quot;,<br>         &quot;versions&quot;: [<br>            &quot;latest-1&quot;<br>         ]<br>      },<br>      {<br>         &quot;browser&quot;: &quot;Firefox&quot;,<br>         &quot;platform&quot;: &quot;Windows 10&quot;,<br>         &quot;versions&quot;: [<br>            &quot;latest-1&quot;<br>         ]<br>      }<br>   ],<br>   &quot;run_settings&quot;: {<br>      &quot;reporter_config_file&quot;: &quot;base_reporter_config.json&quot;,<br>      &quot;build_name&quot;: &quot;Cypress Test&quot;,<br>      &quot;parallels&quot;: 1,<br>      &quot;specs&quot;: &quot;cypress/e2e/tests/checkboxRadio.cy.js&quot;,<br>      &quot;ignore_files&quot;: &quot;&quot;,<br>      &quot;network&quot;: false,<br>      &quot;video&quot;: true,<br>      &quot;headless&quot;: false,<br>      &quot;downloads&quot;: &quot;./cypress/htmlReports/index.html&quot;,<br>      &quot;npm_dependencies&quot;: {<br>         &quot;@cypress/grep&quot;: &quot;^4.1.0&quot;,<br>         &quot;cypress&quot;: &quot;^14.5.1&quot;,<br>         &quot;cypress-downloadfile&quot;: &quot;^1.2.4&quot;<br>      }<br>   },<br>   &quot;tunnel_settings&quot;: {<br>      &quot;tunnel&quot;: false,<br>      &quot;tunnel_name&quot;: null<br>   }<br>}</pre><p>In the above configuration,</p><p>“specs”: “cypress/e2e/tests/checkboxRadio.cy.js”, is to mention testfile name or folder name.</p><p>Step 3 : To Run in local</p><p>&gt; lambdatest-cypress run</p><p>This will run our test in Lambdatest. And view the results in lambdatest,</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*ZSVv-WameHMQ7L8X6bG38Q.png" /></figure><p>Step 4: Create a remote repo in GitLab and push your project.</p><p>Step 5 : Configure CI/CD pipeline variables</p><p>Do not hardcode credentials in your config file. Instead, add them as protected variables in GitLab:</p><p>1. Navigate to your GitLab project <strong>Settings &gt; CI/CD &gt; Variables</strong>.</p><p>2. Add LT_USERNAME (your LambdaTest username).</p><p>3. Add LT_ACCESS_KEY (your LambdaTest access key).</p><p>4. Add npm_config_cache = $CI_PROJECT_DIR/.npm</p><p>5. Add CYPRESS_CACHE_FOLDER = $CI_PROJECT_DIR/cache/Cypress</p><p>Step 6 : Create a Pipeline in GitLab</p><p>Navigate to our project -&gt; Build -&gt; Pipeline -&gt; Create a new Pipeline</p><p>which will create .gitlab.yml file in Pipeline Editor,</p><pre># Global variables<br>variables:<br>  npm_config_cache: &quot;$CI_PROJECT_DIR/.npm&quot;<br>  CYPRESS_CACHE_FOLDER: &quot;$CI_PROJECT_DIR/cache/Cypress&quot;<br>  LT_USERNAME: &quot;$LT_USERNAME&quot;<br>  LT_ACCESS_KEY: &quot;$LT_ACCESS_KEY&quot;<br><br># Cache npm and Cypress binaries<br>cache:<br>  key: ${CI_COMMIT_REF_SLUG}<br>  paths:<br>    - .npm/<br>    - cache/Cypress/<br><br>install:<br>  stage: install<br>  image: cypress/base:18.16.0<br>  script:<br>    - npm ci<br>    - echo &quot;Installing dependencies...&quot;<br>    - rm -rf $CYPRESS_CACHE_FOLDER<br>    - npx cypress install<br>    - chmod +x ./node_modules/.bin/cypress<br>  artifacts:<br>    paths:<br>      - node_modules/<br><br>test:<br>  stage: test<br>  image: cypress/base:18.16.0<br>  dependencies:<br>    - install<br>  script:<br>    - echo &quot;Running Cypress tests on LambdaTest...&quot;<br>    - node -v<br>    - npm -v<br>    # Run Cypress with LambdaTest credentials<br>    # - npx cypress run --env LT_USERNAME=$LT_USERNAME,LT_ACCESS_KEY=$LT_ACCESS_KEY<br>    # - npx cypress run  <br>    - npm install -g lambdatest-cypress-cli # Install the CLI<br>    - lambdatest-cypress run --lambdatest-config lambdatest-config.json # Execute tests</pre><p>Step 7 : Run pipeline job</p><p>Created a Pipeline with two stages,</p><blockquote>Install</blockquote><blockquote>test</blockquote><p>Upon committing these changes, the pipeline job will be triggered automatically. Once the job is completed, it will generate the results in GitLab as well as in LambdaTest as below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*Pfgh20mAXAiiRPsw_B6gXQ.png" /></figure><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*Ooty0fCcoxpCvIud8esZlQ.png" /></figure><p>Conclusion :</p><p>Cypress tests can be seamlessly integrated into a CI/CD pipeline and executed on LambdaTest, enabling validation across diverse environments. This emphasizes both the automation aspect (CI/CD) and the flexibility of running tests in multiple environments through LambdaTest.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=77d54897d77b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[GraphQL API Mocking using Playwright — Part 3]]></title>
            <link>https://anandhik.medium.com/graphql-api-testing-using-playwright-part-3-c53f59954f0c?source=rss-bc7bd0890b9f------2</link>
            <guid isPermaLink="false">https://medium.com/p/c53f59954f0c</guid>
            <category><![CDATA[graphqlmocking]]></category>
            <category><![CDATA[api-playwright]]></category>
            <category><![CDATA[playwrightgraphql]]></category>
            <category><![CDATA[playwrights]]></category>
            <dc:creator><![CDATA[Anandhi K]]></dc:creator>
            <pubDate>Wed, 18 Mar 2026 07:44:40 GMT</pubDate>
            <atom:updated>2026-03-18T08:03:01.474Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>(API Mocking)</strong></p><p>This article is on GraphQL API mocking using Playwright with Typescript.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/644/1*Fm_ByIn46e3g0ijyZVuevg.png" /></figure><p>In this series we are exploring the following key topics:</p><ol><li>GraphQL API Testing using Playwright — Part 1 (Query)</li></ol><p><a href="https://medium.com/@anandhik/graphql-api-testing-using-playwright-part-1-ddc8ba881e00">GraphQL API Testing using Playwright — Part 1 | by Anandhi K | Medium</a></p><p>2. GraphQL API Testing using Playwright — Part 2 (Mutation)</p><p><a href="https://medium.com/@anandhik/graphql-api-testing-using-playwright-part-2-mutation-8c72b5323b2b">GraphQL API Testing using Playwright — Part 2 (Mutation) | by Anandhi K | Medium</a></p><p>3. GraphQL API mocking using Playwright</p><p>In this third part, we will see — GraphQL API mocking using Playwright.</p><p><strong>API Mocking</strong></p><p>API mocking is the practice of simulating the behavior of an actual API by creating fake endpoints or predefined responses. It’s a powerful technique used during development and testing when the real API is unavailable, unstable, or still under construction.</p><p><strong>Why API Mocking?</strong></p><ul><li><strong>Parallel development</strong>: Frontend teams can build and test UI components without waiting for backend APIs to be ready.</li><li><strong>Testing edge cases</strong>: Simulate rare or error-prone scenarios (e.g., timeouts, 500 errors) that are hard to reproduce with a live API.</li><li><strong>Stability</strong>: Avoid flakiness caused by network issues or backend changes during testing.</li><li><strong>Speed</strong>: Reduce dependency on external systems, speeding up test execution and CI/CD pipelines.</li></ul><p><strong>GraphQL API Mocking using Playwright</strong></p><p>API endpoint URL</p><p><a href="https://apollo-fullstack-tutorial.herokuapp.com/graphql"><em>https://apollo-fullstack-tutorial.herokuapp.com/graphql</em></a></p><p>Steps to create this sample,</p><p>1. Create a project in VS Code</p><p>2. Install playwright</p><p>3. Provide below configuration in playwright.confige.ts file,</p><pre>projects: [<br>    //configured for GraphQL API testing<br>    {<br>      name: &#39;api-apollo-service&#39;,<br>      use: {<br>        //Use for Mocking<br>        // baseURL: &#39;https://apollo-fullstack-tutorial.herokuapp.com&#39;,<br>        //Use for UI Fetch<br>        baseURL: &#39;https://apollo-fullstack-tutorial.herokuapp.com/graphql&#39;,<br>        trace: &#39;on&#39;<br>      },<br>      testMatch: /apiTest\/apollo</pre><p>In this example,</p><p>· We set up a Playwright network interceptor for GraphQL POSTs and a small test that triggers a GraphQL request and asserts the mocked response.</p><p>· Intercept requests to any /graphql endpoint, return a canned “launches” response when the operation name is Launches, and verify the client receives that mocked payload.</p><p><strong>Code Walkthrough (section-by-section)</strong></p><p><strong>· queryGetAll </strong>— Holds a GraphQL query string and variables</p><pre>const queryGetAll = {<br>    query: `<br>        query Launches($pageSize: Int, $after: String) {<br>            launches(pageSize: $pageSize, after: $after) {<br>                cursor<br>                hasMore<br>                launches {<br>                id<br>                site<br>                }<br>            }<br>        }`,<br>    variables: { &quot;pageSize&quot;: 2, &quot;after&quot;: &quot; &quot; }<br>};</pre><p><strong>· getQueryOperationName(query) </strong>— Extracts the GraphQL operation name using a regex: matches query &lt;Name&gt; and returns the Name or null.</p><pre>function getQueryOperationName(query: string): string | null {<br>    const match = query.match(/query\s+(\w+)/);<br>    return match ? match[1] : null;<br>}</pre><p><strong>· hasOperationName(query, operationName)</strong> — Helper that returns true when the extracted operation name equals the expected one.</p><pre>function hasOperationName(query: string, operationName: string): boolean {<br>    return (<br>        getQueryOperationName(query) === operationName);<br>}</pre><p><strong>· mockGraphQLRequest(page: Page)</strong></p><p>o Registers a route handler for ‘**/graphql’ (all URLs ending with /graphql).</p><p>o route.request().postDataJSON() reads the POST JSON body (the GraphQL request payload).</p><p>o Extracts the query string and logs it.</p><p>o If the operation name is ‘Launches’, fulfill the request with a mocked JSON response (cursor, hasMore, launches array).</p><p>o Otherwise, calls route.continue() to let the real request go through.</p><pre>export const mockGraphQLRequest = async (page: Page) =&gt; {<br>    console.log(&#39;mockGraphQL......&#39;)<br>    await page.route(&#39;**/graphql&#39;, async (route) =&gt; {<br>        const body = route.request().postDataJSON();<br>        const query = body.query;<br>        console.log(&#39;PostJSON Query&#39;, query)<br>        if (hasOperationName(query, &#39;Launches&#39;)) {<br>            await route.fulfill({<br>                status: 200,<br>                contentType: &#39;application/json&#39;,<br>                body: JSON.stringify({<br>                    data: {<br>                        &quot;launches&quot;: {<br>                            &quot;cursor&quot;: &quot;1604618640&quot;,<br>                            &quot;hasMore&quot;: true,<br>                            &quot;launches&quot;: [<br>                                {<br>                                    &quot;id&quot;: &quot;MR200&quot;,<br>                                    &quot;site&quot;: &quot;KSC&quot;<br>                                },<br>                                {<br>                                    &quot;id&quot;: &quot;MR201&quot;,<br>                                    &quot;site&quot;: &quot;CCAFS&quot;<br>                                }]<br>                        }<br>                    }<br>                }),<br>            });<br>            return;<br>        }<br>        await route.continue();<br>    });<br>};</pre><p><strong>· Test block</strong></p><p>o Creates a new browser page and installs the mock via await mockGraphQLRequest(page).</p><p>o Triggers a GraphQL POST from the page context using page.evaluate(fetch(…)) with a query named Launches.</p><p>o Awaits res.json() returned from the fetch (this should be the mocked payload).</p><p>o Asserts with expect that the returned payload matches the mocked values (cursor, hasMore, launches array).</p><pre>test(&#39;Mocked response in landing page&#39;, async ({ browser }) =&gt; {<br>    const page = await browser.newPage()<br>    await mockGraphQLRequest(page)<br>    // Capture and verify the mocked response<br>    const response = await page.evaluate(async () =&gt; {<br>        const res = await fetch(&#39;https://apollo-fullstack-tutorial.herokuapp.com/graphql&#39;, {<br>            method: &#39;POST&#39;,<br>            headers: { &#39;Content-Type&#39;: &#39;application/json&#39; },<br>            body: JSON.stringify({<br>                query: `<br>                        query Launches {<br>                            launches {<br>                                id<br>                                site<br>                            }<br>                        }`<br>            })<br>        })<br>        return res.json()<br>    })<br><br>    // Assert the mocked response - in case of &#39;launches&#39; query<br>    expect(response.data.launches.cursor).toBe(&#39;1604618640&#39;)<br>    expect(response.data.launches.hasMore).toBe(true)<br>    console.log(&#39;Response data...&#39;, response.data.launches.launches)<br>    expect(response.data.launches.launches).toEqual([<br>        { id: &#39;MR200&#39;, site: &#39;KSC&#39; },<br>        { id: &#39;MR201&#39;, site: &#39;CCAFS&#39; }<br>    ])<br>})</pre><p><strong>Best practices followed</strong></p><p>· Ensure the intercepted request is a POST and postDataJSON() can be null; add guards (if (!body) return route.continue()) to avoid runtime errors.</p><p>· The regex used extracts only named queries of the form query Name; unnamed or differently formatted operations won’t match.</p><p>· page.evaluate runs in the page context.</p><p>· Using ‘**/graphql’ matches many hosts; tighten the pattern if you only want a single endpoint.</p><p>After running, this generates a report as below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RkBbwwkv3fY765fZ4HRIBA.png" /></figure><p>With console output as,</p><p>Response data… [ { id: ‘MR200’, site: ‘KSC’ }, { id: ‘MR201’, site: ‘CCAFS’ } ]</p><p><strong>Conclusion</strong></p><p>Thus, Playwright provides built-in support for mocking GraphQL API responses using route(), following a similar approach to REST API mockings.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=c53f59954f0c" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[GraphQL API Testing using Playwright — Part 2 (Mutation)]]></title>
            <link>https://anandhik.medium.com/graphql-api-testing-using-playwright-part-2-mutation-8c72b5323b2b?source=rss-bc7bd0890b9f------2</link>
            <guid isPermaLink="false">https://medium.com/p/8c72b5323b2b</guid>
            <category><![CDATA[graphqlapiautomation]]></category>
            <category><![CDATA[playwright-test]]></category>
            <category><![CDATA[playwrights]]></category>
            <category><![CDATA[playwrightgraphqlapi]]></category>
            <category><![CDATA[playwright-automation]]></category>
            <dc:creator><![CDATA[Anandhi K]]></dc:creator>
            <pubDate>Sat, 01 Nov 2025 12:26:05 GMT</pubDate>
            <atom:updated>2025-11-01T12:26:05.653Z</atom:updated>
            <content:encoded><![CDATA[<h3>GraphQL API Testing using Playwright — Part 2 (Mutation)</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/644/1*Fm_ByIn46e3g0ijyZVuevg.png" /></figure><p>This series is on testing GraphQL API using Playwright with Typescript. Basics are covered in Part 1.</p><p>In this second part, we’ll see GraphQL API testing by exploring the following key topic:</p><blockquote>Testing GraphQL mutation using Playwright</blockquote><p><strong>GraphQL API Mutation :</strong></p><p>A <strong>GraphQL mutation</strong> is a type of operation used to <strong>create, update, or delete data</strong> on the server — essentially, any action that changes the state of your application.</p><p>Here, we will see 2 scenarios:</p><p>· To book some launch trips</p><p>· To cancel a launch trip</p><p><strong>Scenario 1 — To book some launch trips</strong></p><pre>mutation BookTrips($launchIds: [ID]!) {<br>        bookTrips(launchIds: $launchIds) {<br>                success<br>                message<br>                launches {<br>                site<br>                }<br>            }<br>        }<br><br>mutation CancelTrip($launchId: ID!) {<br>        cancelTrip(launchId: $launchId) {<br>            message<br>            success<br>            launches {<br>            id<br>            site<br>            }<br>        }<br>     }</pre><p>And to execute these mutate operations, we need to generate a token for authorization. Once a token is generated, it should be used as a header.</p><p><strong>Testing GraphQL mutation using Playwright :</strong></p><p>API endpoint URL : <a href="https://apollo-fullstack-tutorial.herokuapp.com/graphql">https://apollo-fullstack-tutorial.herokuapp.com/graphql</a></p><p><strong>Steps to create this sample,</strong></p><p>1. Create a project in VS Code</p><p>2. Install playwright</p><p>3. Provide the below configuration in playwright.confige.ts file.</p><pre>projects: [<br>    //configured for GraphQL API testing<br>    {<br>      name: &#39;api-apollo-service&#39;,<br>      use: {<br>        //Use for Mocking<br>        // baseURL: &#39;https://apollo-fullstack-tutorial.herokuapp.com&#39;,<br>        //Use for UI Fetch<br>        baseURL: &#39;https://apollo-fullstack-tutorial.herokuapp.com/graphql&#39;,<br>        trace: &#39;on&#39;<br>      },<br>      testMatch: /apiTest\/apollo-service\/.*.spec.ts/<br>    },</pre><p>4. Using the Apollo GraphQL server, we can execute a query as below.</p><p>To generate token,</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/652/1*77hn1d3UmXeIfSXJsb7mXA.png" /></figure><p>To execute mutate operation,</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/659/1*UgyqTogML8II3DYa4GzzWA.png" /></figure><p>5. Create a mutate query by passing a valid email id which will generate a token. And use this token as a header. To perform this, include the code below.</p><pre>let token;<br>const headers = {<br>    Authorization: token<br>}<br><br>const loginMutation = `<br>  mutation Login($email: String) {<br>    login(email: $email) {<br>      token<br>    }<br>  }`<br>const loginOptions = {<br>    data: {<br>        query: loginMutation,<br>        variables: { email: &#39;corpdevops@gmail.com&#39; },<br>    },<br>}</pre><p>6. Create a mutate query and its expected response as variables in a file mutateBooking.spec.ts.</p><pre>const mutateBookTrip = {<br>    data: {<br>        query: `mutation BookTrips($launchIds: [ID]!) {<br>        bookTrips(launchIds: $launchIds) {<br>                success<br>                message<br>                launches {<br>                site<br>                }<br>            }<br>        }`,<br>        variables: { &quot;launchIds&quot;: [&quot;110&quot;, &quot;108&quot;] }<br>    },<br>    headers: headers  //passing authorization token<br>}<br><br><br>//Expected response <br>const mulExpResonses = {<br>    &quot;bookTrips&quot;: {<br>        &quot;success&quot;: true,<br>        &quot;message&quot;: &quot;trips booked successfully&quot;,<br>        &quot;launches&quot;: [<br>            {<br>                &quot;site&quot;: &quot;KSC LC 39A&quot;<br>            },<br>            {<br>                &quot;site&quot;: &quot;VAFB SLC 4E &quot;<br>            }<br>        ]<br>    }<br>}</pre><p>7. Write a test to execute this query and validate its expected response.</p><pre>test.beforeEach(async ({ request, baseURL }) =&gt; {<br>    // Login to get token<br>    const loginReq = await request.post(baseURL as string, loginOptions);<br>    expect(loginReq.status()).toBe(200);<br>    const loginRes = await loginReq.json();<br>    token = loginRes[&#39;data&#39;][&#39;login&#39;][&#39;token&#39;];<br>    console.log(&#39;Token.....&#39;, token)<br>    headers.Authorization = token;<br>})<br><br>test(&quot;Book Trips&quot;, async ({ request, baseURL }) =&gt; {<br>    const res = await request.post(baseURL as string, mutateBookTrip);<br>    expect(res.status()).toBe(200);<br>    const resData = await res.json();<br>    console.log(resData.data.bookTrips)<br>    const tripDetail = resData.data.bookTrips<br>    // Assert successful booking and expected site for multiple launchID<br>    expect(resData.data).toEqual(mulExpResonses);<br>    expect(tripDetail.success).toBe(true);<br>    expect(tripDetail.message).toEqual(&#39;trips booked successfully&#39;)<br>    expect(tripDetail.launches[0]).toHaveProperty(&quot;site&quot;, &quot;KSC LC 39A&quot;) //launchID=110<br>    expect(tripDetail.launches[1]).toHaveProperty(&quot;site&quot;, &quot; VAFB SLC 4E<br>&quot;) //launchID=108<br><br>})</pre><p>In the above test, ‘request’ fixture is used to send an API request using the post method. On successful execution, this will send a response object, and we verify this response object for the expected status and response data.</p><p>8. This will generate the below report.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/655/1*m4gOKk1fyfQDR9FCWOtrXA.png" /></figure><p><strong>Scenario 2 — To cancel a launch trip</strong></p><p>To cancel a trip, create a query as below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/637/1*rCpOoNF9-htWi5IYF5aU5g.png" /></figure><p>Create a separate test file and include below code.</p><pre>import { test, expect } from &#39;@playwright/test&#39;<br>const mutateCancelTrip = {<br>    data: {<br>        query: `mutation CancelTrip($launchId: ID!) {<br>        cancelTrip(launchId: $launchId) {<br>            message<br>            success<br>            launches {<br>            id<br>            site<br>            }<br>        }<br>            }`,<br>        variables: { launchId: 110 }<br>    }<br>}<br><br>const expResCT = {<br>    &quot;cancelTrip&quot;: {<br>        &quot;message&quot;: &quot;trip cancelled&quot;,<br>        &quot;success&quot;: true,<br>        &quot;launches&quot;: [<br>            {<br>                &quot;id&quot;: &quot;110&quot;,<br>                &quot;site&quot;: &quot;KSC LC 39A&quot;<br>            }<br>        ]<br>    }<br>}<br><br>test(&quot;MutateTripCancel Test&quot;, async ({ request, baseURL }) =&gt; {<br>    // Initiates GraphQL query &amp; saves the json response<br>    const res = await request.post(baseURL as string, mutateCancelTrip);<br>    const resData = await res.json();<br>    console.log(resData.data.cancelTrip)<br>    const tripDetail = resData.data.cancelTrip<br>    // Asserts response status &amp; response data object<br>    expect(res.status()).toBe(200);<br>    expect(resData[&quot;data&quot;]).toEqual(expResCT);<br>    expect(tripDetail.message).toEqual(&#39;trip cancelled&#39;)<br>    expect(tripDetail.success).toBeTruthy()<br>    expect(tripDetail.launches[0]).toHaveProperty(&quot;site&quot;, &quot;KSC LC 39A&quot;)<br><br>})</pre><p>This will generate the below report.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/725/1*J15MZWyaF9ufDBe1uEuC8w.png" /></figure><p><strong>Conclusion:</strong></p><p>Playwright provides built-in support for testing GraphQL APIs, following a similar approach to REST API CRUD operations.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8c72b5323b2b" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[GraphQL API Testing using Playwright — Part 1]]></title>
            <link>https://anandhik.medium.com/graphql-api-testing-using-playwright-part-1-ddc8ba881e00?source=rss-bc7bd0890b9f------2</link>
            <guid isPermaLink="false">https://medium.com/p/ddc8ba881e00</guid>
            <category><![CDATA[playwright-testing]]></category>
            <category><![CDATA[graphql-api]]></category>
            <category><![CDATA[playwrightgraphql]]></category>
            <category><![CDATA[playwrights]]></category>
            <category><![CDATA[graphql]]></category>
            <dc:creator><![CDATA[Anandhi K]]></dc:creator>
            <pubDate>Sat, 25 Oct 2025 07:46:30 GMT</pubDate>
            <atom:updated>2025-10-25T07:46:30.135Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/644/1*Fm_ByIn46e3g0ijyZVuevg.png" /></figure><h3><strong>GraphQL API Testing using Playwright — Part 1</strong></h3><p>This series is on testing GraphQL API using Playwright with Typescript.</p><p>In this first part, we’ll lay the foundation for effective GraphQL testing by exploring the following key topics:</p><blockquote>1. What is GraphQL?</blockquote><blockquote>2. Advantage over RestAPI</blockquote><blockquote>3. Type of GraphQL Operations</blockquote><blockquote>4. Testing GraphQL queries using Playwright</blockquote><p>This series is designed to automate GraphQL API testing using Playwright with Typescript, Part I will equip you with the essential concepts and practical scaffolds to get started confidently.</p><p><strong>What is GraphQL?</strong></p><p>GraphQL is a query language for your API, and a server-side runtime for executing queries using a type system you define for your data. At its core, GraphQL enables declarative data fetching, where a client can specify exactly what data it needs from an API. Instead of multiple endpoints that return fixed data structures, a GraphQL server only exposes a single endpoint and responds with precisely the data a client asked for. It’s like ordering food from a menu where you can customize your meal down to the last detail.</p><p><strong>Advantages over RestAPI :</strong></p><p>1. <strong>Precise Data Fetching</strong></p><ul><li>Clients can request exactly the data they need — no more, no less.</li><li>Avoids over-fetching (getting unnecessary data) and under-fetching (missing needed data), which is common in REST where endpoints return fixed structures.</li></ul><p>2. <strong>Single Endpoint</strong></p><ul><li>GraphQL uses a single endpoint for all queries and mutations.</li><li>REST typically requires multiple endpoints for different resources, which can lead to redundancy and maintenance overhead.</li></ul><p>3. <strong>Better for Complex Relationships</strong></p><ul><li>Ideal for nested or relational data (e.g., user → posts → comments).</li><li>GraphQL queries can traverse and fetch deeply nested data in one request, whereas REST would require multiple round-trips.</li></ul><p>4. <strong>Strong Typing and Introspection</strong></p><ul><li>GraphQL schemas are strongly typed and self-documenting.</li><li>Tools like GraphiQL or Apollo Studio allow developers to explore the API and auto-generate documentation.</li></ul><p>5. <strong>Improved Developer Experience</strong></p><ul><li>Frontend teams can iterate faster without backend changes.</li><li>Schema evolution is easier with GraphQL’s versionless approach, unlike REST which often requires versioning for changes.</li></ul><p>6. <strong>Efficient for Mobile and Low-Bandwidth Clients</strong></p><ul><li>Mobile apps benefit from reduced payload sizes and fewer requests.</li><li>This makes GraphQL especially attractive for performance-sensitive applications.</li></ul><p>7. <strong>Real-Time Data with Subscriptions</strong></p><ul><li>GraphQL supports subscriptions for real-time updates.</li><li>REST typically relies on polling or third-party solutions for similar functionality.</li></ul><p><strong>Type of GraphQL Operations :</strong></p><blockquote>Queries to get data</blockquote><blockquote>Mutations to change data (create, update, delete)</blockquote><blockquote>Subscriptions to get real-time updates</blockquote><p><strong>Queries :</strong></p><p>A <strong>GraphQL query</strong> is a structured request used to retrieve specific data from a GraphQL API. Unlike REST, where you hit different endpoints for different resources, GraphQL allows you to query <strong>exactly</strong> what you need from a <strong>single endpoint</strong> — nothing more, nothing less.</p><p>The home page can ask for just the book titles and authors with a simple GraphQL query.</p><pre>{<br>  books {<br>    title<br>    author {<br>      name<br>    }<br>  }<br>}</pre><p>The book details page can ask for detailed information about a specific book using another GraphQL query.</p><p>The user profile page can ask for the user’s information and their purchase history with another GraphQL query.</p><p><strong>Mutations :</strong></p><p>A <strong>GraphQL mutation</strong> is a type of operation used to <strong>create, update, or delete data</strong> on the server — essentially, any action that changes the state of your application.</p><pre>mutation {<br>    addBook(title: &quot;The Catcher in the Rye&quot;, author: &quot;J.D. Salinger&quot;) {<br>    title<br>    author {<br>      name<br>    }<br>  }<br>}</pre><p><strong>Subscriptions:</strong></p><p>A <strong>GraphQL subscription</strong> is a powerful feature that enables <strong>real-time communication</strong> between the client and server. Unlike queries and mutations — which are one-time operations — subscriptions maintain a <strong>persistent connection</strong> (usually via WebSockets) so the server can push updates to the client as events occur.</p><p>Subscriptions are ideal for scenarios where data changes frequently and users need live updates, such as:</p><ul><li>Chat messages</li><li>Notifications</li><li>Live scores or analytics</li><li>Collaborative editing</li></ul><p><strong>Testing GraphQL queries using Playwright :</strong></p><p>API endpoint URL: <a href="https://apollo-fullstack-tutorial.herokuapp.com/graphql">https://apollo-fullstack-tutorial.herokuapp.com/graphql</a></p><p><strong>Steps to create this sample,</strong></p><p>1. Create a project in VS Code</p><p>2. Install playwright</p><p>3. Provide below configuration in playwright.confige.ts file,</p><pre>projects: [<br>    //configured for GraphQL API testing<br>    {<br>      name: &#39;api-apollo-service&#39;,<br>      use: {<br>        //Use for Mocking<br>        // baseURL: &#39;https://apollo-fullstack-tutorial.herokuapp.com&#39;,<br>        //Use for UI Fetch<br>        baseURL: &#39;https://apollo-fullstack-tutorial.herokuapp.com/graphql&#39;,<br>        trace: &#39;on&#39;<br>      },<br>      testMatch: /apiTest\/apollo-service\/.*.spec.ts/<br>    },</pre><p>4. Using Apollo GraphQL server, we can execute a query as below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/694/1*6SXAy_fRB7mAC5cSFAw_DA.png" /></figure><p>5. Create the above query and its expected response as variables in a file called getAllQuery.spec.ts.</p><pre>const getAllQuery = {<br>    data: {<br>        query: `query Launches($pageSize: Int, $after: String) {<br>   launches(pageSize: $pageSize, after: $after) {<br>    cursor<br>    hasMore<br>    launches {<br>      site<br>      id<br>    }<br>   }<br>}`,<br>        variables: { &quot;pageSize&quot;: 5, &quot;after&quot;: &quot; &quot; },<br>    }<br>}<br>//Expected response for answers<br>const expectedRes = {<br>    &quot;launches&quot;: {<br>      &quot;cursor&quot;: &quot;1603553460&quot;,<br>      &quot;hasMore&quot;: true,<br>      &quot;launches&quot;: [<br>        {<br>          &quot;site&quot;: &quot;KSC LC 39A&quot;,<br>          &quot;id&quot;: &quot;110&quot;<br>        },<br>        {<br>          &quot;site&quot;: &quot;CCAFS SLC 40&quot;,<br>          &quot;id&quot;: &quot;109&quot;<br>        }<br>      ]<br>    }<br>  }</pre><p>6. Write a test to execute this query and validate its expected response.</p><pre>test(&quot;Query to get Rocket Launches&quot;, async ({ request, baseURL }) =&gt; {<br>    // Initiates GraphQL query &amp; saves the json response<br>    const res = await request.post(baseURL as string, getAllQuery);<br>    const resData = await res.json();<br>    console.log(resData.data.launches.launches)<br>    // Asserts response status &amp; response data object<br>    expect(res.status()).toBe(200);<br>    // Test should PASS if actual response equals expected response<br>    expect(resData[&quot;data&quot;]).toEqual(expectedRes);<br><br>    const explaunches = resData.data.launches.launches;<br>    expect(explaunches).toHaveLength(2)<br>    // expect(explaunches[0]).toHaveProperty(&quot;site&quot;, &quot;KSC LC 39A&quot;)<br>    expect(explaunches[1]).toHaveProperty(&quot;site&quot;, &quot;CCAFS SLC 40&quot;)<br>})</pre><p>In the above test, the ‘request’ fixture is used to send API request using the post method. On successful execution, this will send a response object and we verify this response object for the expected status and response data.</p><p>7. This will generate the below report.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/628/1*wtJt1LPsT8mZ6BkSGSI3jA.png" /></figure><p><strong>Conclusion:</strong></p><p>Playwright provides built-in support for testing GraphQL APIs, following a similar approach to REST API testing workflows.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ddc8ba881e00" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Supercharging Selenium with MCP Server and Claude AI]]></title>
            <link>https://anandhik.medium.com/supercharging-selenium-with-mcp-server-and-claude-ai-42d7e269555a?source=rss-bc7bd0890b9f------2</link>
            <guid isPermaLink="false">https://medium.com/p/42d7e269555a</guid>
            <category><![CDATA[claudeselenium]]></category>
            <category><![CDATA[claude-ai]]></category>
            <category><![CDATA[selenium-mcp]]></category>
            <category><![CDATA[mcpclaude-ai]]></category>
            <category><![CDATA[mcp-server]]></category>
            <dc:creator><![CDATA[Anandhi K]]></dc:creator>
            <pubDate>Thu, 19 Jun 2025 11:40:03 GMT</pubDate>
            <atom:updated>2025-06-24T05:13:08.768Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*C2X6np_C-dfyEl0DkhJtTg.jpeg" /></figure><p>In this article, we’ll explore just how effortless it is to generate Selenium automation scripts in a short amount of time by leveraging the power of the <strong>MCP Server</strong> and <strong>Claude AI</strong>. With minimal setup, this combination opens up a more intuitive and AI-assisted approach to web automation.</p><p>We’ll walk through the following:</p><p>What is MCP?</p><p>How MCP helps in Test Automation?</p><p>Using Claude AI to Generate Selenium Scripts :</p><blockquote><strong>Step 1:</strong> Install Claude AI on your local machine<br> <strong>Step 2:</strong> Integrate Claude AI with the Selenium-compatible MCP Server<br> <strong>Step 3:</strong> Interact with web applications using Claude AI and observe how it interprets web elements<br> <strong>Step 4:</strong> Generate Selenium scripts dynamically through the MCP API<br> <strong>Step 5: E</strong>xecute the generated scripts from Eclipse</blockquote><p><strong>What is MCP?</strong></p><p>MCP (Model Context Protocol Server) — is a middleware component that acts as a bridge between AI agents (like Claude AI) and browser automation tools such as <strong>Selenium WebDriver</strong>. It provides a standardized interface for automating browser interactions, making it easier to script and control web actions programmatically.</p><p>Think of it as the translator that allows AI models to “talk” to browsers: navigating to URLs, clicking buttons, filling out forms, and more — all through a consistent API.</p><p><strong>How it is helpful in Test Automation?</strong></p><p>Using an <strong>MCP Server</strong> in test automation — especially with tools like Selenium — brings a whole new level of intelligence and efficiency to your workflow. Here’s how it helps:</p><p><strong>· AI-Powered Test Execution</strong>: MCP acts as a bridge between AI agents (like Claude or Copilot) and browser automation tools. This allows you to describe test steps in natural language and have them executed programmatically.</p><p><strong>· Faster Debugging &amp; RCA</strong>: It streamlines root cause analysis (RCA) by giving you direct access to logs (network, console, Selenium) and test execution data — right from your IDE.</p><p><strong>· Test Case Generation</strong>: Based on previous executions, MCP can help generate new test cases, making your suite smarter over time.</p><p><strong>· CI/CD Integration</strong>: MCP servers can be integrated into your CI/CD pipelines, enabling scalable, distributed test execution with better resource management.</p><p><strong>· Unified Interface</strong>: Think of it as a “USB-C for automation” — a universal protocol that simplifies how AI tools interact with browsers and test frameworks.</p><p><strong>Using Claude AI to Generate Selenium Scripts :</strong></p><p><strong>Step 1:</strong> Install Claude AI on your local machine</p><p>a) Install Claude.ai in your local from <a href="https://claude.ai/download">Download — Claude</a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/523/1*LAjBlUJvSQAR8Vm9JQlLcQ.png" /></figure><p>b) When prompting to launch application in browser for automation, we get the below message</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/525/1*Rqf9ZrV3dPjpeVMn9BQ52g.png" /></figure><p>It means, any LLM cannot interact with any other tools like browser, local files, data sources etc. To enable Claude.ai to interact with Chrome browser we need to install MCP Server.</p><p>MCP official documentation — <a href="https://github.com/modelcontextprotocol">Model Context Protocol</a></p><p><strong>Step 2:</strong> Integrate Claude AI with the Selenium-compatible MCP Server</p><p>To configure Selenium MCP Server, Goto</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/529/1*POhUeE7wRjT8abe49isi-Q.png" /></figure><p>a) Developer -&gt; Edit Config — which will navigate to local claude folder where we have claude_destop_config.json file.</p><p>b) MCP Selenium Server Config — <a href="https://github.com/angiejones/mcp-selenium">angiejones/mcp-selenium: An MCP implementation for Selenium WebDriver</a></p><p>This MCP server can be configured either with Goose Desktop or Claude desktop.</p><p>c) Now, we configure Claude desktop app.</p><pre>Copy the below content<br>{<br>    &quot;mcpServers&quot;: {<br>    &quot;selenium&quot;: {<br>    &quot;command&quot;: &quot;npx&quot;,<br>    &quot;args&quot;: [&quot;-y&quot;, &quot;@angiejones/mcp-selenium&quot;]<br>    }<br>  }<br>}</pre><p>d) Paste above config into claude_destop_config.json file.</p><p>To ensure whether the configuration is complete, verify by navigating to File -&gt; Edit -&gt; Developer</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/533/1*0fiu-Bz6EHjr7qqDUiRVpQ.png" /></figure><p>Also under tools, will display list of available options.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/530/1*MgCWrRO1fpHWJHyVhA9IpA.png" /></figure><p><strong>Step 3:</strong> Interact with web applications using Claude AI and observe how it interprets web elements</p><p>Use the prompt in Claude,</p><p><em>&gt; open chrome browser and launch </em><a href="https://www.saucedemo.com/"><em>https://www.saucedemo.com</em></a></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/562/1*NArGAuYBVzZd0p6Bnh2qRA.png" /></figure><p>This will open browser and launch the given URL.</p><p>Next, prompt is</p><p><em>&gt; And login with standard_user and secret_sauce</em></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/543/1*Zx2YWxj0bQY8MR_Jcs6N8A.png" /></figure><p>Finally, login into application from Chrome browser.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/559/1*N7AGeA647FjnrTKGbSjgKA.png" /></figure><p><strong>Step 4:</strong> Generate Selenium scripts dynamically through the MCP API</p><p>For the above scenario, we are going to generate Selenium testscript using TestNG framework.</p><p>Let the prompt be,</p><p><em>&gt; Generate the above steps in Selenium using Java and TestNG.</em></p><p>This will generate a code as below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/549/1*qnsAv5hYHbE4ZncyRDGsKw.png" /></figure><p>This has generated 2 tests :</p><p>1. Verify login page elements are present</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/575/1*8dfoy-ClqyIebqMJPAWMRA.png" /></figure><p>2. Login into Saucedemo with valid credentials</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/589/1*whjH_4gDWr3VZb05BQ3Jgw.png" /></figure><p><strong>Step 5:</strong> Execute genereated scripts from Eclipse</p><p>i. Claude also generated instructions to configure Maven project :</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/511/1*aMN0kjwX2KBMUqGEunS4MA.png" /></figure><p>ii. Create a TestNG class called SauceDemoLoginTest.</p><p>iii. Copy and paste the above generated code and run as TestNG test from eclipse.</p><pre>package mcpServerGenerated;<br><br><br>import org.openqa.selenium.By;<br>import org.openqa.selenium.WebDriver;<br>import org.openqa.selenium.WebElement;<br>import org.openqa.selenium.chrome.ChromeDriver;<br>import org.openqa.selenium.support.ui.WebDriverWait;<br>import org.openqa.selenium.support.ui.ExpectedConditions;<br>import org.testng.Assert;<br>import org.testng.annotations.AfterMethod;<br>import org.testng.annotations.BeforeMethod;<br>import org.testng.annotations.Test;<br>import java.time.Duration;<br><br>public class SauceDemoLoginTest {<br>    <br>    private WebDriver driver;<br>    private WebDriverWait wait;<br>    <br>    // Test data<br>    private static final String BASE_URL = &quot;https://www.saucedemo.com&quot;;<br>    private static final String USERNAME = &quot;standard_user&quot;;<br>    private static final String PASSWORD = &quot;secret_sauce&quot;;<br>    <br>    @BeforeMethod<br>    public void setUp() {<br>        // Initialize ChromeDriver<br>        driver = new ChromeDriver();<br>        <br>        // Initialize WebDriverWait with 10 seconds timeout<br>        wait = new WebDriverWait(driver, Duration.ofSeconds(10));<br>        <br>        // Maximize browser window<br>        driver.manage().window().maximize();<br>        <br>        // Navigate to SauceDemo website<br>        driver.get(BASE_URL);<br>    }<br>    <br>    @Test(description = &quot;Login to SauceDemo with valid credentials&quot;)<br>    public void testLoginWithValidCredentials() {<br>        // Find and enter username<br>        WebElement usernameField = wait.until(<br>            ExpectedConditions.elementToBeClickable(By.id(&quot;user-name&quot;))<br>        );<br>        usernameField.clear();<br>        usernameField.sendKeys(USERNAME);<br>        <br>        // Find and enter password<br>        WebElement passwordField = driver.findElement(By.id(&quot;password&quot;));<br>        passwordField.clear();<br>        passwordField.sendKeys(PASSWORD);<br>        <br>        // Click login button<br>        WebElement loginButton = driver.findElement(By.id(&quot;login-button&quot;));<br>        loginButton.click();<br>        <br>        // Verify successful login by checking if we&#39;re redirected to products page<br>        wait.until(ExpectedConditions.urlContains(&quot;inventory.html&quot;));<br>        <br>        // Assert that we&#39;re on the products page<br>        String currentUrl = driver.getCurrentUrl();<br>        Assert.assertTrue(currentUrl.contains(&quot;inventory.html&quot;), <br>            &quot;Login failed - not redirected to inventory page&quot;);<br>        <br>        // Verify products page title is displayed<br>        WebElement productsTitle = wait.until(<br>            ExpectedConditions.visibilityOfElementLocated(<br>                By.className(&quot;title&quot;)<br>            )<br>        );<br>        Assert.assertEquals(productsTitle.getText(), &quot;Products&quot;, <br>            &quot;Products page title not found&quot;);<br>        <br>        System.out.println(&quot;Login test passed successfully!&quot;);<br>        System.out.println(&quot;Current URL: &quot; + currentUrl);<br>    }<br>    <br>    @Test(description = &quot;Verify login page elements are present&quot;)<br>    public void testLoginPageElements() {<br>        // Verify username field is present<br>        WebElement usernameField = wait.until(<br>            ExpectedConditions.presenceOfElementLocated(By.id(&quot;user-name&quot;))<br>        );<br>        Assert.assertTrue(usernameField.isDisplayed(), &quot;Username field not displayed&quot;);<br>        <br>        // Verify password field is present<br>        WebElement passwordField = driver.findElement(By.id(&quot;password&quot;));<br>        Assert.assertTrue(passwordField.isDisplayed(), &quot;Password field not displayed&quot;);<br>        <br>        // Verify login button is present<br>        WebElement loginButton = driver.findElement(By.id(&quot;login-button&quot;));<br>        Assert.assertTrue(loginButton.isDisplayed(), &quot;Login button not displayed&quot;);<br>        Assert.assertTrue(loginButton.isEnabled(), &quot;Login button not enabled&quot;);<br>        <br>        System.out.println(&quot;Login page elements verification passed!&quot;);<br>    }<br>    <br>    @AfterMethod<br>    public void tearDown() {<br>        // Close the browser<br>        if (driver != null) {<br>            driver.quit();<br>        }<br>    }<br>}</pre><p>iv. After successfully executing the script, TestNG has generated the results as below.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/520/1*ey4zVqZp3Obcg-Yq1TKzKw.png" /></figure><p><strong>Conclusion :</strong></p><p>Finally, we have an option to automate our Test Automation using MCP Servers.</p><p>Incorporating MCP servers into your Selenium test automation strategy offers a scalable and efficient way to manage distributed test execution. By centralizing orchestration, MCP enhances test performance, simplifies resource allocation, and ensures smoother CI/CD integration. It’s a smart step toward robust, enterprise-grade automation.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=42d7e269555a" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Keyword Library for Robot Framework]]></title>
            <link>https://anandhik.medium.com/keyword-library-for-robot-framework-8ae996b04036?source=rss-bc7bd0890b9f------2</link>
            <guid isPermaLink="false">https://medium.com/p/8ae996b04036</guid>
            <category><![CDATA[robot-framework]]></category>
            <category><![CDATA[robot-framework-test]]></category>
            <category><![CDATA[customlibrary]]></category>
            <category><![CDATA[own-python-libraries]]></category>
            <category><![CDATA[robot-framework-library]]></category>
            <dc:creator><![CDATA[Anandhi K]]></dc:creator>
            <pubDate>Tue, 22 Apr 2025 09:03:11 GMT</pubDate>
            <atom:updated>2025-04-22T09:03:11.652Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*oMsQH5HKiei1zz3XSQiwwQ.jpeg" /></figure><p>This article is about creating a custom keyword library in Python and using the same in Robot Framework as a keyword.</p><p>Robot Framework is a generic open source framework for automating tasks and can be used to perform automated software testing with a focus on acceptance testing. In addition to creating custom keywords in Robot and using external Robot standard libraries, it is very easy to create your own Robot keyword libraries in languages such as Python to encapsulate sensitive code, for example. In general, anything that can be implemented as program code in Python can also be implemented as a Robot keyword library.</p><p>Prerequisites</p><blockquote>Python 3.9 or above</blockquote><blockquote>Robot Framework 7.2 or above</blockquote><p><a href="https://tech.bertelsmann.com/en/blog/articles/workshop-create-a-robot-framework-keyword-library-with-python">Workshop: Create a Robot Framework Keyword Library with Python</a></p><p>Project Structure :</p><p>tests — contains our final Robot Framework test suite that uses our Python Robot Keyword library.</p><p>DemoCalculatorLibrary — is the directory for our Python robot library. It is recommended to always choose the directory name analogous to the library name.</p><p>Our library supports the following feature set:</p><p>Input parameters:</p><ul><li>Operator(+,-,*, %…)</li><li>Two Operands(a, b)</li></ul><p>Output parameter:</p><ul><li>Calculated value</li></ul><p>Step 1 : Create a project folder with the name SampleCustomLibrary.</p><p>Step 2 : Create our Robot library with Python</p><p>First, we create the basic framework of our library and create our two Python files under a folder DemoCalculatorLibrary.</p><p><strong>__init__.py</strong> (for the creation of a later Python package)</p><pre>from .DemoCalculatorLibrary import DemoCalculatorLibrary</pre><p><strong>DemoCalculatorLibrary.py </strong>(this is the library itself):</p><p>Here we first start with our standard includes. I use the standard Python logger for local tests and for the integration into the Robot Framework the necessary Robot.</p><p>Create a file with the name DemoCalculatorLibrary.py under DemoCalculatorLibrary folder.</p><pre>from robot.api.deco import library, keyword, not_keyword<br>from datetime import datetime<br>import logging<br><br>logging.basicConfig(<br>    level=logging.DEBUG, format=&quot;%(asctime)s %(module)s -%(levelname)s- %(message)s&quot;<br>)<br>logger = logging.getLogger(__name__)<br>__version__ = &quot;0.1&quot;<br>__author__ = &quot;Anandhi K&quot;</pre><p>This is created as a Python class. We thus create our class and create or initialize our class variables. As we want to turn this code into a Robot library, we declare the necessary code via a decorator which is part of the <a href="https://robot-framework.readthedocs.io/en/latest/autodoc/robot.api.html#module-robot.api.deco">robot.api.deco</a> library. For our keyword library, we specify a global scope and deactivate the automatic creation of robot keywords from our Python library, so that only those Python functions with a @keyword decorator are exposed as Robot keywords.</p><pre>@library(scope=&quot;GLOBAL&quot;, auto_keywords=True)<br>class DemoCalculatorLibrary:<br>    #Default parameter settings<br>    DEFAULT_OPERATOR = None<br>    DEFAULT_PARAMETER_ONE = 0<br>    DEFAULT_PARAMETER_TWO = 0<br><br>    # Class-internal parameters<br>    __demo_operator = None<br>    __demo_parameter_one = None<br>    __demo_parameter_two = None<br><br>And this class can be initialized with a constructor, <br>def __init__(<br>        self,<br>        demo_parameter_one: int =  DEFAULT_PARAMETER_ONE,<br>        demo_parameter_two: int =  DEFAULT_PARAMETER_TWO,<br>        demo_operator: str = DEFAULT_OPERATOR,<br>    ):<br>        self.__demo_operator = demo_operator<br>        self.__demo_parameter_one = demo_parameter_one<br>        self.__demo_parameter_two = demo_parameter_two<br></pre><p>Now we create the Python specific “getter”/”setter” methods. For the “setter” methods we check if a value has been passed.</p><pre>#<br>    # Python &quot;getter&quot; methods<br>    #<br>    @property<br>    def demo_operator(self):<br>        return self.__demo_operator<br><br>    @property<br>    def demo_parameter_one(self):<br>        return self.__demo_parameter_one<br><br>    @property<br>    def demo_parameter_two(self):<br>        return self.__demo_parameter_two<br><br> #<br>    # Python &quot;setter&quot; methods<br>    #<br>    @demo_operator.setter<br>    def demo_operator(self, demo_operator: str):<br>        if not demo_operator:<br>            raise ValueError(&quot;No &#39;Operator&#39; value has been specified&quot;)<br>        self.__demo_operator = demo_operator<br><br>    @demo_parameter_one.setter<br>    def demo_parameter_one(self, demo_parameter_one: int):<br>        if not demo_parameter_one:<br>            raise ValueError(&quot;No &#39;Parameter One&#39; value has been specified&quot;)<br>        self.__demo_parameter_one = demo_parameter_one<br><br>    @demo_parameter_two.setter<br>    def demo_parameter_two(self, demo_parameter_two: int):<br>        if not demo_parameter_two:<br>            raise ValueError(&quot;No &#39;Parameter Two&#39; value has been specified&quot;)<br>        self.__demo_parameter_two = demo_parameter_two</pre><p>Now, we will create getter / setter methods in Robot Framework to invoke these methods.</p><pre> #<br>    # Robot &quot;getter&quot; methods<br>    #<br>    @keyword(&quot;Get Operator&quot;)<br>    def get_demo_operator(self):<br>        return self.demo_operator<br><br>    @keyword(&quot;Get Parameter One&quot;)<br>    def get_demo_parameter_one(self):<br>        return self.demo_parameter_one<br><br>    @keyword(&quot;Get Parameter Two&quot;)<br>    def get_demo_parameter_two(self):<br>        return self.demo_parameter_two<br>      <br>          #<br>    # Robot &quot;setter&quot; methods<br>    #<br>    @keyword(&quot;Set Operator&quot;)<br>    def set_demo_operator(self, operator: str = None):<br>        logging.debug(msg=&quot;Setting Demo &#39;operator&#39;&quot;)<br>        self.demo_operator = operator<br><br>    @keyword(&quot;Set Parameter One&quot;)<br>    def set_demo_parameter_one(self, parameter_one: int = 0):<br>        logging.debug(msg=&quot;Setting Demo &#39;parameter one&#39;&quot;)<br>        self.demo_parameter_one = parameter_one<br><br>    @keyword(&quot;Set Parameter Two&quot;)<br>    def set_demo_parameter_two(self, parameter_two: int = 0):<br>        logging.debug(msg=&quot;Setting Demo &#39;parameter two&#39;&quot;)<br>        self.demo_parameter_two = parameter_two</pre><p>Now, create a method in Python to perform arithmetic calculations. Then we create a separate keyword, which determines a type of operator and perform corresponding function. The actual determination is done by a subroutine, which we explicitly do not publish as a robot keyword in our library (@not_keyword decorator).</p><pre>    #<br>    # An internal function that we deliberately exclude<br>    # from being promoted to a Robot Framework keyword<br>    #<br>    @not_keyword<br>    def get_calculated_value(self, operator: str, parameter_one: int, parameter_two: int):<br>        logging.debug(msg=&quot;Calculating value&quot;)<br>        retval = None<br><br>        if operator == &quot;+&quot;:<br>            retval = parameter_one + parameter_two<br>        elif operator == &quot;-&quot;:<br>            retval = parameter_one - parameter_two<br>        elif operator == &quot;*&quot;:<br>            retval = parameter_one * parameter_two  <br>        elif operator == &quot;/&quot;:<br>            if parameter_two == 0:<br>                raise ValueError(&quot;Division by zero is not allowed&quot;)<br>            retval = parameter_one / parameter_two<br>        elif operator == &quot;%&quot;:<br>            if parameter_two == 0:<br>                raise ValueError(&quot;Division by zero is not allowed&quot;)<br>            retval = parameter_one % parameter_two<br>        elif operator == &quot;**&quot;:<br>            retval = parameter_one ** parameter_two<br>        elif operator == &quot;//&quot;:<br>           if parameter_two == 0:<br>                raise ValueError(&quot;Division by zero is not allowed&quot;)<br>            retval = parameter_one // parameter_two<br>        elif operator == &quot;sqrt&quot;:<br>            if parameter_one &lt; 0:<br>                raise ValueError(&quot;Square root of negative number is not allowed&quot;)<br>            retval = parameter_one ** (1 / parameter_two)<br>        else:<br>             raise ValueError(f&quot;Unknown operator &#39;{operator}&#39;&quot;)<br>        return retval</pre><pre>  #<br>    # Our demo function which will return a calculated value back to the user<br>    #<br>    @keyword(&quot;Evaluate Expression&quot;)<br>    def evaluate_expression(self, operator: str = None, parameter_one: int = None, parameter_two: int = None):<br>        logging.debug(msg=&quot;Calculate value&quot;)<br> <br>        # Get our operator. The operator can be taken from either this keyword or from the class.<br>        # At least one operator needs to be set_demo_operator<br>        if not self.demo_operator and not operator:<br>            raise ValueError(&quot;No operator specified&quot;)<br><br>        op = operator if operator else self.demo_operator<br><br>        # Get our parameters. The parameters can be taken from either this keyword or from the class.<br>        # At least one parameter needs to be set_demo_parameter_one and set_demo_parameter_two<br>        param_one = parameter_one if parameter_one else self.demo_parameter_one<br>        param_two = parameter_two if parameter_two else self.demo_parameter_two<br><br>        return_value = self.get_calculated_value(op, param_one, param_two)<br>        return return_value <br><br><br>    if __name__ == &quot;__main__&quot;:<br>        pass</pre><p>Our Robot Framework library is ready and we can now include it in a Robot Framework test case. Here, we use new keyword(Evaluate Expression) from our library</p><p>Step 3 : Create our Robot Test Case</p><pre>*** Settings ***<br> <br>Library    DemoCalculatorLibrary<br>Library    OperatingSystem<br><br>*** Variables ***<br>${NUM1}    10    <br>${NUM2}    20    <br><br>*** Test Cases ***<br>Addition Test<br>    [Documentation]    This test case verifies the addition of two numbers.<br>    ${result}=    Evaluate Expression     +      ${NUM1}    ${NUM2}<br>    Should Be Equal As Numbers    ${result}    30<br><br>Subtraction Test<br>     ${result}=    Evaluate Expression     -      ${NUM2}    ${NUM1}<br>    Should Be Equal As Numbers    ${result}    10</pre><p>To run our Robot Test Case, perform the below configuration.</p><p>To be able to run our robot test without publishing our Robot library to the <a href="https://www.pypi.org/">PyPi </a>index, we need to point the PYTHONPATH variable to the directory where our Python library is located:</p><pre>Windows:<br>set PYTHONPATH=F:\Anandhi\workspace\robotframework\SampleCustomLibrary\DemoCalculatorLibrary\<br><br><br>Now let&#39;s run our Robot test suite <br>  &gt;   Robot DemoCalculatorTest.robot</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*1QJWgChGnh32LqrikXJu0A.png" /></figure><p>When we pass some decimal number to this keyword, our test will fail.</p><pre>*** Variables ***<br>${NUM1}    10    <br>${NUM2}    20   <br>${NUM3}    10.5    <br>${NUM4}    20  <br><br>*** Test Cases ***<br>Addition Test<br>    [Documentation]    This test case verifies the addition of two numbers.<br>    ${result}=    Evaluate Expression     +      ${NUM1}    ${NUM2}<br>    Should Be Equal As Numbers    ${result}    30<br><br>Subtraction Test<br>     ${result}=    Evaluate Expression     -      ${NUM4}    ${NUM3}<br>    Should Be Equal As Numbers    ${result}    10</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*TDpawBK9CyQKppwNIZCpTw.png" /></figure><p>Conclusion</p><p>Thus, we have created our own keyword library in Python for Robot Framework.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8ae996b04036" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Robot Framework Browser Library]]></title>
            <link>https://anandhik.medium.com/robot-framework-browser-library-480f01cfa92e?source=rss-bc7bd0890b9f------2</link>
            <guid isPermaLink="false">https://medium.com/p/480f01cfa92e</guid>
            <category><![CDATA[playwrights]]></category>
            <category><![CDATA[robot-framework-browser]]></category>
            <category><![CDATA[robot-framework]]></category>
            <category><![CDATA[browser-library]]></category>
            <dc:creator><![CDATA[Anandhi K]]></dc:creator>
            <pubDate>Tue, 01 Apr 2025 07:25:50 GMT</pubDate>
            <atom:updated>2025-04-01T07:25:50.291Z</atom:updated>
            <content:encoded><![CDATA[<figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*N031Xf6mwa68dzN5gpkPNQ.png" /></figure><p><strong>(Powered by Playwright)</strong></p><p>Impressive… it’s a keyword-driven adaptation of Playwright!.</p><p>Robot Framework Browser is fast! With its direct API to the browsers, designed for high performance testing, fast feedback is now only limited by the speed of the test object.</p><p>The browser library automatically waits for the elements of the page and interacts with them when they are ready for interaction.</p><p>No more flaky tests that are sometimes PASS and sometimes FAIL are drastically reduced and confidence in test automation increases!</p><p>So, as testers focus will be on finding real errors instead of fixing tests…</p><p>To install this library,</p><blockquote>pip install robotframework-browser</blockquote><p>Introduction</p><p>Browser library works with three different layers that build on each other: <strong>Browser</strong>, <strong>Context</strong> and <strong>Page</strong>.</p><p><strong>Browsers</strong></p><p>A <strong>browser</strong> can be started with one of the three different engines Chromium, Firefox or Webkit.</p><p><strong>Supported Browsers</strong></p><figure><img alt="" src="https://cdn-images-1.medium.com/max/698/1*1r8zH5ttMLOpAR90jnY91w.png" /></figure><p>A browser process is started headless (without a GUI) by default. Run a new browser with specified arguments if a browser with a GUI is requested or if a proxy has to be configured. A browser process can contain several contexts.</p><p><strong>Contexts</strong></p><p>A <strong>context</strong> corresponds to a set of independent incognito pages in a browser that share cookies, sessions or profile settings. Pages in two separate contexts do not share cookies, sessions or profile settings. Compared to Selenium, these do <strong>not</strong> require their own browser process. To get a clean environment, a test can just open a new context. The context layer is useful e.g. for testing different user sessions on the same webpage without opening a whole new browser context.</p><p><strong>Pages</strong></p><p>A <strong>page</strong> does contain the content of the loaded web site and has a browsing history. Pages and browser tabs are the same.</p><p>The <a href="https://marketsquare.github.io/robotframework-browser/Browser.html#Open%20Browser"><em>Open Browser</em></a> keyword opens a new browser, a new context and a new page. This keyword is useful for quick experiments or debugging sessions.</p><p>When a <em>new page</em> is called without an open browser, <a href="https://marketsquare.github.io/robotframework-browser/Browser.html#New%20Browser"><em>New Browser</em></a> and <a href="https://marketsquare.github.io/robotframework-browser/Browser.html#New%20Context"><em>New Context</em></a> are executed with default values first.</p><p>Each Browser, Context and Page has a unique ID with which they can be addressed.</p><p><strong>Selectors</strong></p><p>The default selector strategy is CSS selector.</p><p>Browser library supports the same selector strategies as the underlying Playwright node module: xpath, css, id and text. The strategy can either be explicitly specified with a prefix or the strategy can be implicit.</p><p>A major advantage of Browser is that multiple selector engines can be used within one selector. It is possible to mix XPath, CSS and Text selectors while selecting a single element.</p><p>One of the major advantages here is that selectors are strings that consists of one or more clauses separated by &gt;&gt; token, e.g. clause1 &gt;&gt; clause2 &gt;&gt; clause3. When multiple clauses are present, next one is queried relative to the previous one’s result. Browser library supports concatenation of different selectors separated by &gt;&gt;.</p><p><strong>Accessing WebElements</strong></p><p>For eg:</p><pre>text=Hello &gt;&gt; xpath=../.. &gt;&gt; css=.select_button</pre><p>The above selector contains — text, xpath and css.</p><p><strong>Accessing iFrames</strong></p><p>By default, selector chains do not cross frame boundaries. It means that a simple CSS selector is not able to select an element located inside an iframe or a frameset. For this use case, there is a special selector &gt;&gt;&gt; which can be used to combine a selector for the frame and a selector for an element inside a frame.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*HFVKGDGt5gUycJw1Z1VbFg.png" /></figure><pre>To select an element from an iFrame,<br> ${label}    Get Text    div &gt; h3<br>     Log     The text is : ${label}<br>    ${txtFromFrame}     Get Text   iframe#mce_0_ifr &gt;&gt;&gt; id=tinymce       <br> Log     ${txtFromFrame}</pre><p><strong>Accessing Shadow DOMs</strong></p><p>Playwright and so also Browser are able to do automatic piercing of Shadow DOMs and therefore are the best automation technology when working with WebComponents.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/940/1*OVQjP53DPpT6RjHRmtODIA.png" /></figure><p>Both input and button elements are inside a Shadow Tree. To access them</p><pre>New Page    http://uitestingplayground.com/shadowdom<br>${DOMElem}    Get Element    guid-generator &gt; input#editField<br>     Type Text       ${DOMElem}    Welcome<br>     Click       guid-generator &gt; button#buttonGenerate<br>     ${DomText}    Get Text    guid-generator &gt; input#editField<br>Log To Console    message=${DomText}</pre><p><strong>Assertions</strong></p><p>Keywords that accept arguments assertion_operator and assertion_expected can optionally assert that a specified condition holds. Keywords will return the value even when the assertion is performed by the keyword.</p><p>For eg :</p><pre>${strTxt} Get Text   .flash.error    contains    invalid </pre><p>The above code is fetching the text from .flash.error and also verify whether it contains text ‘invalid’.</p><p>Assert will retry and fail only after a specified timeout.</p><p>Some assertion operators are,</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/915/1*GP1rTJpORqWBtItk2BjSCQ.png" /></figure><p><strong>Reference</strong></p><p><a href="https://marketsquare.github.io/robotframework-browser/Browser.html">Browser</a></p><p><strong>Conclusion</strong></p><p>The Browser library offers an extensive set of keywords for interacting with browser elements, serving as a replacement for SeleniumLibrary. Compared to SeleniumLibrary, it delivers superior performance due to its built-in automatic wait mechanism, which enhances efficiency.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=480f01cfa92e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[ChainTest Report Generation using Selenium + TestNG]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://anandhik.medium.com/chaintest-report-generation-using-selenium-testng-b0a3b9cd4c51?source=rss-bc7bd0890b9f------2"><img src="https://cdn-images-1.medium.com/max/1019/1*ff7sy6J1YwcrJFqkepAldw.png" width="1019"></a></p><p class="medium-feed-snippet">In this article, we will generate a ChainTest Report using Selenium and TestNG. Say goodbye to ExtentReport as we transition to the more&#x2026;</p><p class="medium-feed-link"><a href="https://anandhik.medium.com/chaintest-report-generation-using-selenium-testng-b0a3b9cd4c51?source=rss-bc7bd0890b9f------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://anandhik.medium.com/chaintest-report-generation-using-selenium-testng-b0a3b9cd4c51?source=rss-bc7bd0890b9f------2</link>
            <guid isPermaLink="false">https://medium.com/p/b0a3b9cd4c51</guid>
            <category><![CDATA[reporting]]></category>
            <category><![CDATA[chain-test]]></category>
            <category><![CDATA[chaintestselenium]]></category>
            <category><![CDATA[extentreport]]></category>
            <dc:creator><![CDATA[Anandhi K]]></dc:creator>
            <pubDate>Wed, 22 Jan 2025 15:46:49 GMT</pubDate>
            <atom:updated>2025-06-19T06:49:49.242Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Mobile Automation using Appium 2.0 Part 1-Long Press]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://anandhik.medium.com/mobile-automation-using-appium-2-0-part-1-long-press-a2c314b39d9d?source=rss-bc7bd0890b9f------2"><img src="https://cdn-images-1.medium.com/max/633/1*TCCO1jEVCnLtStJrHRJ75Q.png" width="633"></a></p><p class="medium-feed-snippet">Appium 1.8 introduced a list of mobile gestures as part of the W3C WebDriver specification&apos;s Actions. Appium supports a variety of mobile&#x2026;</p><p class="medium-feed-link"><a href="https://anandhik.medium.com/mobile-automation-using-appium-2-0-part-1-long-press-a2c314b39d9d?source=rss-bc7bd0890b9f------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://anandhik.medium.com/mobile-automation-using-appium-2-0-part-1-long-press-a2c314b39d9d?source=rss-bc7bd0890b9f------2</link>
            <guid isPermaLink="false">https://medium.com/p/a2c314b39d9d</guid>
            <category><![CDATA[mobile-action]]></category>
            <category><![CDATA[appiummobileactions]]></category>
            <category><![CDATA[appiumlongpress]]></category>
            <category><![CDATA[javaclient9]]></category>
            <category><![CDATA[appium-java]]></category>
            <dc:creator><![CDATA[Anandhi K]]></dc:creator>
            <pubDate>Fri, 18 Oct 2024 11:10:38 GMT</pubDate>
            <atom:updated>2024-10-18T11:10:38.754Z</atom:updated>
        </item>
        <item>
            <title><![CDATA[Playwright tests with AI — ZeroStep]]></title>
            <description><![CDATA[<div class="medium-feed-item"><p class="medium-feed-image"><a href="https://anandhik.medium.com/playwright-tests-with-ai-zerostep-6bfd7c31fb8f?source=rss-bc7bd0890b9f------2"><img src="https://cdn-images-1.medium.com/max/1280/1*8pbH5oXtvASv7sPwTx7mtg.jpeg" width="1280"></a></p><p class="medium-feed-snippet">ZeroStep is an npm package introduced to flavour Playwright tests with AI power by using GPT3.5 and GPT 4.0 to make the script more&#x2026;</p><p class="medium-feed-link"><a href="https://anandhik.medium.com/playwright-tests-with-ai-zerostep-6bfd7c31fb8f?source=rss-bc7bd0890b9f------2">Continue reading on Medium »</a></p></div>]]></description>
            <link>https://anandhik.medium.com/playwright-tests-with-ai-zerostep-6bfd7c31fb8f?source=rss-bc7bd0890b9f------2</link>
            <guid isPermaLink="false">https://medium.com/p/6bfd7c31fb8f</guid>
            <category><![CDATA[playwright-ai]]></category>
            <category><![CDATA[aizerostep]]></category>
            <category><![CDATA[playwright-test]]></category>
            <category><![CDATA[playwright-automation]]></category>
            <dc:creator><![CDATA[Anandhi K]]></dc:creator>
            <pubDate>Sat, 21 Sep 2024 13:36:53 GMT</pubDate>
            <atom:updated>2024-09-21T13:43:02.777Z</atom:updated>
        </item>
    </channel>
</rss>