<?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 Steve Zebib on Medium]]></title>
        <description><![CDATA[Stories by Steve Zebib on Medium]]></description>
        <link>https://medium.com/@mzebib?source=rss-f49c260596a5------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/1*LDX2vfrR1aouEXYQg1rTww.png</url>
            <title>Stories by Steve Zebib on Medium</title>
            <link>https://medium.com/@mzebib?source=rss-f49c260596a5------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Mon, 16 Nov 2026 10:54:36 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@mzebib/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[Oracle AI Agent Studio: How to Invoke an Agent Asynchronously]]></title>
            <link>https://medium.com/oracle-saas-paas/oracle-ai-agent-studio-how-to-invoke-an-agent-asynchronously-a481723179b8?source=rss-f49c260596a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/a481723179b8</guid>
            <category><![CDATA[oracle-ai-agent-studio]]></category>
            <category><![CDATA[agentic-applications]]></category>
            <category><![CDATA[oracle-fusion]]></category>
            <category><![CDATA[oracle]]></category>
            <category><![CDATA[ai-agent]]></category>
            <dc:creator><![CDATA[Steve Zebib]]></dc:creator>
            <pubDate>Fri, 11 Sep 2026 16:58:33 GMT</pubDate>
            <atom:updated>2026-09-11T19:50:07.608Z</atom:updated>
            <content:encoded><![CDATA[<h3>Overview</h3><p>This article shows how to invoke an agent from Oracle AI Agent Studio asynchronously using REST APIs.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*IXvIa9r7fv0sRhF1cncikA.png" /></figure><h3><strong>Prerequisites</strong></h3><h4>Parameters</h4><ul><li><strong><em>SERVER_URL</em></strong> — (e.g. <a href="https://servername.fa.us2.oraclecloud.com/">https://servername.fa.us2.oraclecloud.com</a>)</li><li><strong><em>WORKFLOW_CODE</em></strong> — Agent workflow code (e.g. SAMPLE_FUSION_AGENT)</li><li><strong><em>ACCESS_TOKEN</em></strong> — Bearer token from the AI Agent Studio Chat API request (see below)</li></ul><h4>Getting the Access Token</h4><p>In <strong>AI Agent Studio</strong>, invoke the agent and locate the Chat API call in your browser’s <strong>Developer Tools → Network</strong> tab.</p><p>For example:</p><pre>https://servername.fa.us2.oraclecloud.com/api/fusion-ai/orchestrator/agent/v2/SAMPLE_FUSION_AGENT/chat</pre><p>Copy the <strong>Bearer token</strong> from the Authorization request header and use it as ACCESS_TOKEN in the APIs below.</p><h3>Steps</h3><h4>Step 1: Invoke Async</h4><p>Use the invokeAsync API to start the agent asynchronously.</p><pre>POST {{SERVER_URL}}/api/fusion-ai/orchestrator/agent/v2/{{WORKFLOW_CODE}}/invokeAsync</pre><p><strong>Example:</strong></p><pre>POST https://servername.fa.us2.oraclecloud.com/api/fusion-ai/orchestrator/agent/v2/SAMPLE_FUSION_AGENT/invokeAsync</pre><p>Headers:</p><pre>Authorization: Bearer {{ACCESS_TOKEN}}<br>Content-Type: application/json</pre><p>Payload:</p><pre>{<br>  &quot;conversational&quot;: true,<br>  &quot;message&quot;: &quot;tell me about Oracle AI Agent Studio&quot;,<br>  &quot;invocationMode&quot;: &quot;ADMIN&quot;<br>}</pre><p>Response:</p><pre>{<br>  &quot;jobId&quot;: &quot;6c59af92-808e-406a-b2b8-b9f5069ff0d3&quot;,<br>  &quot;conversationId&quot;: &quot;24e7723f-b95b-4f21-ba42-d26fed40a587&quot;,<br>  &quot;status&quot;: &quot;RUNNING&quot;<br>}</pre><p>Save the jobId. It is used in the next step to check the agent execution status.</p><h4>Step 2: Get Status</h4><p>Use the jobId from Step 1 to check the status and retrieve the agent response.</p><pre>GET {{SERVER_URL}}/api/fusion-ai/orchestrator/agent/v2/{{WORKFLOW_CODE}}/status/{{JOB_ID}}?invocationMode=ADMIN</pre><p><strong>Example</strong></p><pre>GET https://servername.fa.us2.oraclecloud.com/api/fusion-ai/orchestrator/agent/v2/SAMPLE_FUSION_AGENT/status/6c59af92-808e-406a-b2b8-b9f5069ff0d3?invocationMode=ADMIN</pre><p>Headers:</p><pre>Authorization: Bearer {{ACCESS_TOKEN}}</pre><p>Response:</p><p>If the agent is still running:</p><pre>{<br>  &quot;jobId&quot;: &quot;6c59af92-808e-406a-b2b8-b9f5069ff0d3&quot;,<br>  &quot;status&quot;: &quot;RUNNING&quot;,<br>  &quot;output&quot;: null<br>}</pre><p>When the agent completes:</p><pre>{<br>  &quot;jobId&quot;: &quot;6c59af92-808e-406a-b2b8-b9f5069ff0d3&quot;,<br>  &quot;status&quot;: &quot;COMPLETE&quot;,<br>  &quot;output&quot;: &quot;Oracle AI Agent Studio...&quot;<br>}</pre><p>If the status is RUNNING, call the status API again. When the status is COMPLETE, the agent response is available in the output field.</p><h3>Summary</h3><p>The asynchronous invocation flow requires two API calls:</p><pre>1. POST /invokeAsync<br>        ↓<br>   Get JOB_ID<br>        ↓<br>2. GET /status/{JOB_ID}<br>        ↓<br>   RUNNING → Check again<br>        ↓<br>   COMPLETE → Read output</pre><p>The first request starts the agent and returns a jobId. The second request uses that jobId to monitor the execution and retrieve the final response.</p><h3>References</h3><p><a href="https://docs.oracle.com/en/cloud/saas/fusion-ai/26c/aiaas/index.html">Oracle AI Agent Studio</a></p><p><a href="https://docs.oracle.com/en/cloud/saas/applications-common/26c/farca/api-fusion-ai-agent-team.html">Oracle Fusion Cloud Applications — Agent Team REST API</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=a481723179b8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/oracle-saas-paas/oracle-ai-agent-studio-how-to-invoke-an-agent-asynchronously-a481723179b8">Oracle AI Agent Studio: How to Invoke an Agent Asynchronously</a> was originally published in <a href="https://medium.com/oracle-saas-paas">oracle-saas-paas</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[Agentic Apps — Oracle AI Agent Studio]]></title>
            <link>https://medium.com/oracle-saas-paas/agentic-apps-oracle-ai-agent-studio-85d577816f2c?source=rss-f49c260596a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/85d577816f2c</guid>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[agentic-applications]]></category>
            <category><![CDATA[oracle-ai-agent-studio]]></category>
            <category><![CDATA[oracle]]></category>
            <category><![CDATA[oracle-fusion]]></category>
            <dc:creator><![CDATA[Steve Zebib]]></dc:creator>
            <pubDate>Fri, 22 May 2026 15:07:41 GMT</pubDate>
            <atom:updated>2026-05-22T19:53:51.854Z</atom:updated>
            <content:encoded><![CDATA[<h3>Agentic Apps — Oracle AI Agent Studio</h3><p><strong><em>DISCLAIMER: </em></strong><em>The views expressed in this story are my own and do not necessarily reflect the views of Oracle.</em></p><h3>Overview</h3><p>Agentic apps combine multiple AI agents into intelligent business applications that help organizations analyze information, support decision-making, and automate actions. Each agent is specialized for a specific domain, such as sales, finance, or inventory, allowing the application to deliver more accurate insights, focused reasoning, and better business outcomes.</p><p>Built with Oracle AI Agent Studio, they combine: AI agents, enterprise data, business workflows, and GenAI capabilities.</p><p>Unlike traditional dashboards or assistants, agentic apps are proactive and action-oriented. They identify risks and opportunities in real time, prioritize what matters most, recommend actions, and help users move work forward instead of simply displaying data.</p><p><strong>Key Capabilities:</strong></p><ul><li>Proactive alerts and recommendations</li><li>Context-aware decision support</li><li>Intelligent workflow coordination</li><li>Action-oriented business insights</li><li>Scalable multi-agent collaboration</li></ul><p><strong>Components:</strong></p><ul><li><strong>Information Displays</strong> — AI-generated charts, tables, lists, and visualizations that provide real-time business insights.</li><li><strong>Advisors</strong> — AI agents that answer questions independently or collaboratively across multiple domains.</li><li><strong>Actions</strong> — AI-generated recommendations and decision points that help users take business actions with approval workflows.</li><li><strong>Communications</strong> — AI-assisted outbound messages, such as emails or Slack notifications, generated from predefined templates.</li></ul><p>Together, these capabilities help businesses create intelligent applications that improve productivity, automate workflows, and support faster, smarter decision-making.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*CaDWA6nt16ZJKpKCyjpdmA.png" /><figcaption>Anatomy of an Agentic App (Source: <a href="https://docs.oracle.com/en/cloud/saas/fusion-ai/aiaas/anatomy-of-an-agentic-app.html">https://docs.oracle.com/en/cloud/saas/fusion-ai/aiaas/anatomy-of-an-agentic-app.html</a>)</figcaption></figure><p>1 — <strong>Ask Oracle Advisor</strong>:<em> Lets you ask natural language questions against the set of agents included in the app.</em></p><p>2—<strong>Summary Section</strong>:<em> Displays the most critical information across all the domains in the app’s agents. The summary is typically the first content users read when opening the app.</em></p><p>3 — <strong>Agents</strong>:<em> The example agentic app includes four agents, each represented by a tile on the dashboard. An agent can provide its own headline to convey the most important information for the app user. Users can zoom in on any agent to keep focus on that agent. Any subsequent questions, actions, and communications are then scoped to the selected agent.</em></p><p>4 — <strong>Widget</strong>:<em> A widget generated by an agent from the widget library. Each widget has space for the agent to provide its own insights, and capabilities that the user can choose from. Widgets can also have agent-specific actions that the user can act upon.</em></p><p>5 — <strong>Priority Actions</strong>:<em> App-specific critical information that require the user’s immediate attention and provide options for the user to take action.</em></p><p>6 — <strong>Communications</strong>:<em> Outbound message suggestions that may be predefined by the app or generated by the LLM. It can start a flow allowing the user to construct the content and send the communication.</em></p><h3>Features</h3><ul><li><strong>AI Agents</strong> — Build intelligent AI agents</li><li><strong>Low-Code Tools</strong> — Create apps with less coding</li><li><strong>Workflow Automation</strong> — Automate business tasks</li><li><strong>GenAI Support</strong> — Use AI and LLM capabilities</li><li><strong>Data Integration</strong> — Connect enterprise data and systems</li><li><strong>Oracle Fusion Integration</strong> — Work with Oracle Fusion Applications</li><li><strong>Third-Party Connections</strong> — Integrate external APIs and tools</li><li><strong>Smart Recommendations</strong> — Deliver AI-driven insights and actions</li><li><strong>Security Controls</strong> — Enterprise security and governance</li><li><strong>Monitoring Tools</strong> — Track agent activity and performance</li><li><strong>Scalable Deployment</strong> — Deploy across the enterprise</li></ul><h3>Architecture</h3><p>Oracle AI Agent Studio connects AI agents, enterprise data, workflows, and business applications into one intelligent platform.</p><p><strong>User Layer</strong></p><ul><li>Oracle Fusion Applications</li><li>Dashboards</li><li>Mobile and web apps</li><li>AI assistants</li></ul><p><strong>AI Agent Layer</strong></p><ul><li>Automate tasks</li><li>Coordinate workflows</li><li>Support business decisions</li></ul><p><strong>GenAI Layer</strong></p><ul><li>AI-generated insights</li><li>Recommendations</li><li>Summaries</li><li>Natural language interactions</li></ul><p><strong>Workflow Layer</strong></p><ul><li>Business processes</li><li>Approvals</li><li>Tasks</li><li>Workflow actions</li></ul><p><strong>Data &amp; Integration Layer</strong></p><ul><li>Oracle Fusion Applications</li><li>Enterprise systems</li><li>APIs</li><li>Third-party applications</li></ul><p><strong>Security Layer</strong></p><ul><li>Access control</li><li>Data security</li><li>Governance</li><li>Monitoring and compliance</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xdR76mVlNoB5dL9LDC9MqA.png" /><figcaption>Oracle AI Agent Studio — Agentic Apps Architecture Diagram</figcaption></figure><h3>Configuration</h3><ul><li><strong>Use Workflow Agent Teams</strong> — Workflow agent teams provide structured inputs and predictable workflows.</li><li><strong>Define Clear Descriptions</strong> — Clearly describe each agent team’s purpose and specialization so the app can identify the right team for each request.</li><li><strong>Build Smaller Specialized Teams</strong> — Create focused, topic-specific agent teams that can be combined within the agentic app when needed.</li><li><strong>Keep Workflows Lightweight</strong> — Reduce unnecessary LLM and agent nodes to improve performance and minimize delays.</li><li><strong>Pass User Context Explicitly</strong> — Conversation history and user input are not automatically shared with internal nodes. Use variables or code nodes to pass required context, such as $context.$system.$inputMessage.</li><li><strong>End with an Agent or LLM Node</strong> — Finishing the workflow with an agent or LLM node helps improve response handling and reduce output streaming issues.</li></ul><p>See <a href="https://docs.oracle.com/en/cloud/saas/fusion-ai/aiaas/overview-of-agentic-apps.html#12-Build-Agentic-Apps-Using-Agents-from-AI-Agent-Studio-12">Build Agentic Apps Using Agents from AI Agent Studio</a> for more details on how to create and configure agentic apps.</p><h3>Fusion Agentic Apps</h3><p>Fusion Agentic Applications help organizations improve operations across Oracle Fusion Cloud Applications, including CX, HCM, SCM, and ERP, by using AI-driven automation, intelligent workflows, and coordinated AI agents to reduce manual work, accelerate business operations, deliver better customer and employee experiences, and enable smarter AI-driven decision-making.</p><h4>CX</h4><p><strong>Capabilities</strong></p><ul><li>Sales insights and opportunity recommendations</li><li>Customer engagement automation</li><li>Cross-sell and upsell suggestions</li><li>Lead and renewal management</li><li>Customer service coordination</li></ul><p><strong>Use Cases</strong></p><ul><li>Improve sales pipeline visibility</li><li>Increase upsell and cross-sell opportunities</li><li>Automate lead follow-up and engagement</li><li>Improve customer retention and renewals</li><li>Enhance customer service experiences</li></ul><h4>HCM</h4><p><strong>Capabilities</strong></p><ul><li>Workforce scheduling</li><li>Payroll and HR workflow automation</li><li>Talent management</li><li>Employee onboarding and support</li><li>HR operations coordination</li></ul><p><strong>Use Cases</strong></p><ul><li>Reduce payroll and scheduling issues</li><li>Automate employee onboarding workflows</li><li>Improve workforce productivity</li><li>Streamline HR operations</li><li>Enhance employee experiences</li></ul><h4><strong>SCM</strong></h4><p><strong>Capabilities</strong></p><ul><li>Supplier and sourcing coordination</li><li>Inventory and logistics monitoring</li><li>Procurement automation</li><li>Compliance tracking</li><li>Supply chain optimization</li></ul><p><strong>Use Cases</strong></p><ul><li>Improve supplier collaboration</li><li>Reduce sourcing and operational costs</li><li>Optimize inventory management</li><li>Streamline procurement workflows</li><li>Increase supply chain efficiency</li></ul><h4>ERP</h4><p><strong>Capabilities</strong></p><ul><li>Financial workflow automation</li><li>Cash collection management</li><li>Invoice and expense processing</li><li>Budgeting and forecasting support</li><li>Financial reporting and analytics</li></ul><p><strong>Use Cases</strong></p><ul><li>Accelerate financial operations</li><li>Improve cash flow management</li><li>Reduce manual finance tasks</li><li>Improve operational visibility</li><li>Support faster business decision-making</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*DkuaA3mHEUYTz2V4Ka1O2g.png" /><figcaption>Sample Fusion Agentic App — Sales Command Center</figcaption></figure><h3>References</h3><p><a href="https://docs.oracle.com/en/cloud/saas/fusion-ai/aiaas/overview.html">Oracle AI Agent Studio</a></p><p><a href="https://www.oracle.com/applications/fusion-ai/ai-agents/">Oracle AI Agents for Fusion Applications</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=85d577816f2c" width="1" height="1" alt=""><hr><p><a href="https://medium.com/oracle-saas-paas/agentic-apps-oracle-ai-agent-studio-85d577816f2c">Agentic Apps — Oracle AI Agent Studio</a> was originally published in <a href="https://medium.com/oracle-saas-paas">oracle-saas-paas</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AI Agent Studio for Oracle Fusion Cloud Applications]]></title>
            <link>https://medium.com/oracle-saas-paas/ai-agent-studio-for-oracle-fusion-cloud-applications-d8a578651903?source=rss-f49c260596a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/d8a578651903</guid>
            <category><![CDATA[ai-agent-development]]></category>
            <category><![CDATA[oracle]]></category>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[oracle-fusion]]></category>
            <dc:creator><![CDATA[Steve Zebib]]></dc:creator>
            <pubDate>Wed, 24 Sep 2025 19:05:07 GMT</pubDate>
            <atom:updated>2026-08-06T17:27:22.984Z</atom:updated>
            <content:encoded><![CDATA[<p><strong><em>DISCLAIMER: </em></strong><em>The views expressed in this story are my own and do not necessarily reflect the views of Oracle.</em></p><h3>Overview</h3><p>AI is changing how users interact with enterprise applications. AI agents can answer questions, retrieve business data, automate business processes, and perform actions across Oracle Fusion Cloud Applications.</p><p>As organizations continue to adopt AI, intelligent agents help improve productivity, streamline business processes, and enhance the user experience.</p><p>Let’s explore Oracle AI Agent Studio, its key features, and the core components used to build AI agents.</p><h3>What is Oracle AI Agent Studio?</h3><p>Oracle AI Agent Studio is a design-time environment within Oracle Fusion Cloud Applications that enables customers and partners to create, configure, extend, deploy, and manage AI agents.</p><p>AI Agent Studio provides reusable components and built-in capabilities that help agents understand user requests, retrieve business information, perform business actions, and interact with Oracle Fusion Cloud Applications and external systems.</p><p>It supports both single-agent and multi-agent solutions, making it easy to build AI agents for everything from simple business tasks to complex business workflows.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*fhtV9nPWwLuV26W_p2cSLQ.png" /></figure><h3>Key Features</h3><p>Oracle AI Agent Studio provides a comprehensive set of capabilities for creating, extending, deploying, and managing AI agents, including:</p><ul><li><strong>Low-code agent development</strong> for creating and configuring AI agents.</li><li><strong>Native Oracle Fusion integration</strong> with direct access to Fusion APIs, business data, application security, and business logic.</li><li><strong>Extensibility</strong> using prebuilt agents and templates to customize existing agents or build new ones for specific business scenarios.</li><li><strong>Multi-agent workflows</strong> that coordinate AI agents and human participants using checkpoints and approvals to support complex, multi-step business processes.</li><li><strong>Retrieval-Augmented Generation (RAG)</strong> using uploaded documents to provide context-aware responses.</li><li><strong>Integration with external REST APIs</strong> to connect with third-party applications and services.</li><li><strong>Built-in email capabilities</strong> for sending notifications and automated communications.</li><li><strong>Deep linking</strong> to Oracle Fusion Application pages for seamless navigation.</li><li><strong>Agent testing and validation</strong> before deployment.</li></ul><h3>Core Components</h3><figure><img alt="" src="https://cdn-images-1.medium.com/max/1023/1*X100R_vetef059grhoudWQ.png" /><figcaption>Source: <a href="https://docs.oracle.com/en/cloud/saas/fusion-ai/aiaas/components-of-ai-agent-studio.html#PREFACE_FA">https://docs.oracle.com/en/cloud/saas/fusion-ai/aiaas/components-of-ai-agent-studio.html#PREFACE_FA</a></figcaption></figure><h4>Agents</h4><p>An <strong>Agent</strong> is an AI-powered component that performs a specific business task. It uses prompts, <strong>Tools</strong>, and optional <strong>Topics</strong> to understand user requests, retrieve information, and perform actions.</p><p>An <strong>Agent</strong>:</p><ul><li>Must have one or more <strong>Tools</strong></li><li>Can have zero or more <strong>Topics</strong></li></ul><p>Agents are designed to perform a specific responsibility. For example, one Agent may retrieve customer information, while another creates service requests or generates reports.</p><h4>Topics</h4><p>A <strong>Topic</strong> defines the subject area or domain for an <strong>Agent</strong>. It contains detailed instructions that guide how an <strong>Agent</strong> interprets requests, makes decisions, and invokes its <strong>Tools</strong>.</p><p><strong>Topics</strong> are optional and can be reused across multiple <strong>Agents</strong>, making it easier to maintain consistent behavior across AI solutions.</p><h4>Tools</h4><p>A <strong>Tool</strong> provides an <strong>Agent</strong> with the capability to retrieve information, perform actions, and interact with Oracle Fusion Applications and external systems.</p><p>Every <strong>Agent</strong> must have at least one <strong>Tool</strong> and can use multiple <strong>Tools</strong> to extend its capabilities and support more complex business tasks and workflows.</p><p>The following <strong>Tool</strong> types are available:</p><p><strong>Business Object Tool</strong></p><p>The <strong>Business Object Tool</strong> enables an <strong>Agent</strong> to interact with Oracle Fusion Cloud Application business objects, such as accounts, contacts, leads, and opportunities.</p><p>Before an <strong>Agent</strong> can use a business object, the business object definition must be created and associated with a <strong>Business Object Tool</strong>.</p><p>Key capabilities include:</p><ul><li>Retrieving Oracle Fusion business data</li><li>Performing CRUD operations using Fusion APIs (GET, POST, and PATCH)</li></ul><p><strong>Deep Link Tool</strong></p><p>The <strong>Deep Link Tool</strong> generates URLs that navigate users directly to specific pages within Oracle Fusion Applications.</p><p>Key capabilities include:</p><ul><li>Generating links to Oracle Fusion Application pages</li><li>Supporting user-specific URL parameters</li></ul><p><strong>Document Tool</strong></p><p>The <strong>Document Tool</strong> enables an <strong>Agent</strong> to answer questions using information contained in uploaded documents through Retrieval-Augmented Generation (RAG).</p><p>Supported document formats include:</p><ul><li>.pdf</li><li>.txt</li><li>.md</li></ul><blockquote><strong>Note:</strong> After uploading documents, run the <strong>Process Agent Document</strong> scheduled process to index the documents before they can be used by an <strong>Agent</strong>.</blockquote><p><strong>Email Tool</strong></p><p>The <strong>Email Tool</strong> enables an <strong>Agent</strong> to send email notifications and other automated communications.</p><p>Key capabilities include:</p><ul><li>Sending emails</li><li>Using predefined email templates</li></ul><p><strong>External REST Tool</strong></p><p>The <strong>External REST Tool</strong> enables an <strong>Agent</strong> to communicate with external applications and services by invoking REST APIs.</p><p>Key capabilities include:</p><ul><li>Calling external REST endpoints</li><li>Supporting HTTP operations such as GET, POST, and PATCH</li></ul><h4>Agent Teams</h4><p>An <strong>Agent Team</strong> is a collection of one or more <strong>Agents</strong> that collaborate to accomplish a business process or workflow.</p><p>An <strong>Agent Team</strong> can consist of:</p><ul><li><strong>A single Agent</strong>, when one Agent is sufficient to handle the entire use case.</li><li><strong>Multiple specialized Agents</strong>, when different tasks require distinct capabilities or expertise.</li></ul><p><strong>Agent Teams</strong> support the following patterns:</p><ul><li><strong>Workflow Pattern</strong> — <strong>Agents</strong> execute in a predefined sequence, with the output of one <strong>Agent</strong> serving as the input to the next until the workflow is complete.</li><li><strong>Supervisor–Worker Pattern</strong> — A supervisor <strong>Agent</strong> analyzes the request, delegates work to one or more specialized worker <strong>Agents</strong>, and coordinates their responses to produce the final result.</li></ul><h3>Common Use Cases</h3><p>Oracle AI Agent Studio can be used to build AI agents across Oracle Fusion Cloud Applications, including <strong>Customer Experience (CX)</strong>, <strong>Human Capital Management (HCM)</strong>, <strong>Enterprise Resource Planning (ERP)</strong>, and <strong>Enterprise Performance Management (EPM)</strong>.</p><p>Common use cases include:</p><ul><li><strong>CX</strong> — Retrieve customer information, manage leads and opportunities, create service requests, and assist sales and customer service teams.</li><li><strong>HCM</strong> — Answer employee questions, automate HR processes, assist with recruiting and onboarding, and retrieve employee information.</li><li><strong>ERP</strong> — Create and update business records, automate approvals, manage supplier and procurement processes, and streamline financial operations.</li><li><strong>EPM</strong> — Support planning, budgeting, forecasting, financial analysis, and reporting processes.</li></ul><p>Additional scenarios include:</p><ul><li>Answering questions using enterprise documents with Retrieval-Augmented Generation (RAG).</li><li>Sending email notifications and automated communications.</li><li>Integrating Oracle Fusion Applications with external systems through REST APIs.</li><li>Coordinating multiple AI agents to automate end-to-end business processes.</li></ul><h3>Conclusion</h3><p>Oracle AI Agent Studio provides a flexible, low-code framework for building AI-powered business solutions within Oracle Fusion Applications. By combining reusable <strong>Agents</strong>, <strong>Tools</strong>, <strong>Topics</strong>, and <strong>Agent Teams</strong>, organizations can develop intelligent assistants that retrieve information, automate business processes, and integrate with enterprise applications.</p><p>Whether you’re building a simple agent to answer questions or designing a team of specialized agents to orchestrate complex workflows, Oracle AI Agent Studio offers the foundational components needed to create scalable, maintainable, and reusable AI solutions.</p><h3>References</h3><p><a href="https://www.oracle.com/applications/fusion-ai/ai-agents/">Oracle AI Agents for Fusion Applications</a></p><p><a href="https://docs.oracle.com/en/cloud/saas/fusion-ai/aiaas/overview.html">Oracle AI Agent Studio</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=d8a578651903" width="1" height="1" alt=""><hr><p><a href="https://medium.com/oracle-saas-paas/ai-agent-studio-for-oracle-fusion-cloud-applications-d8a578651903">AI Agent Studio for Oracle Fusion Cloud Applications</a> was originally published in <a href="https://medium.com/oracle-saas-paas">oracle-saas-paas</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AutoGen — Multi-Agent System Integrated with OpenAI File Search]]></title>
            <link>https://medium.com/oracle-saas-paas/autogen-multi-agent-system-integrated-with-openai-file-search-ce395c4929db?source=rss-f49c260596a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/ce395c4929db</guid>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[agentic-system]]></category>
            <category><![CDATA[autogen]]></category>
            <category><![CDATA[oracle-cloud]]></category>
            <category><![CDATA[openai]]></category>
            <dc:creator><![CDATA[Steve Zebib]]></dc:creator>
            <pubDate>Fri, 25 Jul 2025 14:25:12 GMT</pubDate>
            <atom:updated>2025-07-25T16:23:56.073Z</atom:updated>
            <content:encoded><![CDATA[<h3>AutoGen — Multi-Agent System Integrated with OpenAI File Search</h3><h3>Overview</h3><p>This article demonstrates how to build a multi-agent system with AutoGen, leveraging OpenAI’s file search capabilities.</p><p>Files uploaded to OpenAI Storage can be accessed by the model either directly via file inputs or through the file search tool.</p><p>File Inputs<strong>:</strong></p><ul><li>Files specifically PDF are used as inputs and are referenced by file IDs in requests to the model.</li><li>This is useful for tasks like summarization, where files are processed once and then discarded.</li><li>Reference: <a href="https://platform.openai.com/docs/guides/pdf-files">OpenAI File Inputs</a></li></ul><p>File Search Tool:</p><ul><li>The tool searches the content of uploaded files to enrich the context of its generated responses.</li><li>Files are stored in a vector store, and the tool is used in requests by referencing the vector store IDs.</li><li>This is useful for various use cases including long-term and reusable knowledge base for different tasks.</li><li>Reference: <a href="https://platform.openai.com/docs/guides/tools-file-search">OpenAI File Search</a></li></ul><h3>Setup</h3><h4>OpenAI</h4><ul><li><a href="https://platform.openai.com/api-keys">OpenAI API Key</a></li><li>OpenAI Storage Access (Files and Vector Stores)</li></ul><h4>Environment</h4><p>The following configuration is used in this example:</p><ul><li><a href="https://docs.anaconda.com/miniconda/">Conda for Mac</a></li><li>Python 3.10</li></ul><p>The following is a script to setup the env and install packages:</p><pre>conda create -n autogenDev python=3.10<br>conda activate autogenDev<br>pip install &#39;autogen-agentchat==0.6.4&#39;<br>pip install &#39;autogen-ext[openai]==0.6.4&#39;<br>pip install openai</pre><h3>Sample</h3><p>This example is a multi-agent system that accepts user prompts and file uploads, generating responses derived from the file content.</p><p><strong>Agents:</strong></p><ul><li>Supervisor Agent: Manages task allocation to worker agents and consolidates their results into a unified final response.</li><li>Worker Agent (File Store): Handles OpenAI storage operations, including uploading, deleting files, and managing vector stores.</li><li>Worker Agent (Content Generator): Generates task-driven content based on the provided files.</li></ul><p>Create a Python script <strong>sample.py </strong>and copy/paste the following:</p><pre><br>import asyncio<br>import json<br>import requests<br>from io import BytesIO<br>from typing import List<br>from dataclasses import dataclass<br>from enum import Enum<br><br>from autogen_core import AgentId, MessageContext, RoutedAgent, SingleThreadedAgentRuntime, message_handler<br>from autogen_core import SingleThreadedAgentRuntime<br>from autogen_core.models import ChatCompletionClient, SystemMessage, UserMessage<br>from autogen_ext.models.openai import OpenAIChatCompletionClient<br><br>from openai import OpenAI, AsyncOpenAI<br><br>OPEN_AI_API_KEY = &quot;&lt;OPEN_AI_KEY&gt;&quot;<br>OPEN_AI_MODEL = &quot;gpt-4o&quot;<br><br>USER_TASK=&quot;Generate a document with title, overview, and detailed summary.&quot;<br>SUPERVISOR_AGENT_ID = &quot;Supervisor_Agent&quot;<br>SUPERVISOR_AGENT_PROMPT = &quot;&quot;&quot;You are AI assistant acting as a director. Analyze all responses provided by the worker agents to generate a comprehensive response. \n\n##Acceptance Criteria: - Use only the content generated by the worker agents. - Do not change the content. - Do not assume and speculate.&quot;&quot;&quot;<br>FILE_STORE_WORKER_AGENT_ID = &quot;File_Store_Worker_Agent&quot;<br>CONTENT_GENERATOR_WORKER_AGENT_ID = &quot;Content_Generator_Worker_Agent&quot;<br>CONTENT_GENERATOR_WORKER_AGENT_PROMPT = &quot;&quot;&quot;You are an AI assistant specialized in writing content. \n\n##Acceptance Criteria: - Content must be derived from the provided files. - Only use the provided files. - Do not assume and speculate.&quot;&quot;&quot;<br>CONTENT_GENERATOR_WORKER_AGENT_TASKS = [&quot;Generate a brief overview.&quot;, &quot;Generate a detailed summary.&quot;]<br>USE_VECTOR_STORE = True<br>VECTOR_STORE_NAME = &quot;VS-Sample&quot;<br><br><br>@dataclass<br>class UserTask:<br>    task: str<br><br>@dataclass<br>class FileActionType(Enum):<br>    UPLOAD = &quot;UPLOAD&quot;<br>    DELETE = &quot;DELETE&quot;<br><br>@dataclass<br>class FileStoreMetadata:<br>    file_paths: List[str]<br>    file_ids: List[str]<br>    vector_store_id: str<br><br>@dataclass<br>class FileStoreTask:<br>    action_type: str<br>    file_store_metadata: any<br><br>@dataclass<br>class FileStoreResponse:<br>    file_store_metadata: any<br><br>@dataclass<br>class ContentTask:<br>    agent_id: str<br>    task: str<br><br>@dataclass<br>class ContentResponse:<br>    agent_id: str<br>    text: str<br><br>@dataclass<br>class FinalResult:<br>    result: str<br><br><br>class SupervisorAgent(RoutedAgent):<br>    def __init__(<br>        self,<br>        runtime: any,<br>        model_client: ChatCompletionClient,<br>        open_ai_client: any,<br>        agent_id: str,<br>        prompt: str,<br>        file_store_metadata: FileStoreMetadata,<br>    ) -&gt; None:<br>        super().__init__(description=agent_id)<br>        self._runtime = runtime<br>        self._model_client = model_client<br>        self._open_ai_client = open_ai_client<br>        self._agent_id = agent_id<br>        self._prompt = prompt<br>        self._file_store_metadata = file_store_metadata<br><br>    @message_handler<br>    async def handle_task(self, message: UserTask, ctx: MessageContext) -&gt; FinalResult:<br>        print(f&quot;{&#39;-&#39;*80}\n[{self._agent_id}]\nUser Task: {message.task}&quot;)<br><br><br>        # File Store Worker Agent<br>        print(f&quot;{&#39;-&#39;*80}\n[{self._agent_id}]\nUpload Files&quot;)<br><br>        await FileStoreWorkerAgent.register(<br>            self._runtime,<br>            FILE_STORE_WORKER_AGENT_ID,<br>            lambda: FileStoreWorkerAgent(<br>                agent_id=FILE_STORE_WORKER_AGENT_ID,<br>                open_ai_client=self._open_ai_client<br>            ),<br>        )<br><br>        file_store_worker_agent_id = AgentId(FILE_STORE_WORKER_AGENT_ID, f&quot;{self.id.key}/worker&quot;)<br>        file_store_worker_agent_task = FileStoreTask(action_type=FileActionType.UPLOAD.value, file_store_metadata=self._file_store_metadata)<br>            <br>        print(f&quot;\n[{self._agent_id}]\n{FILE_STORE_WORKER_AGENT_ID} Task: Upload Files&quot;)<br><br>        file_store_worker_agent_response = await self.send_message(file_store_worker_agent_task, file_store_worker_agent_id)<br>        file_store_metadata = file_store_worker_agent_response.file_store_metadata<br><br>        print(f&quot;\n[{self._agent_id}]\n{FILE_STORE_WORKER_AGENT_ID} Response:\nFiles Uploaded: {file_store_metadata}&quot;)<br><br><br>        # Content Generator Worker Agents<br>        print(f&quot;{&#39;-&#39;*80}\n[{self._agent_id}]\nGenerate Content&quot;)<br><br>        content_generator_worker_agents = []<br>        for index, task in enumerate(CONTENT_GENERATOR_WORKER_AGENT_TASKS):<br>            worker_agent_id = f&quot;{CONTENT_GENERATOR_WORKER_AGENT_ID}_{index+1}&quot;<br>            worker_task = task<br><br>            await ContentGeneratorWorkerAgent.register(<br>                self._runtime,<br>                worker_agent_id,<br>                lambda: ContentGeneratorWorkerAgent(<br>                    agent_id=worker_agent_id,<br>                    open_ai_client=self._open_ai_client,<br>                    prompt=CONTENT_GENERATOR_WORKER_AGENT_PROMPT,<br>                    file_store_metadata=file_store_metadata<br>                ),<br>            )<br><br>            print(f&quot;\n[{self._agent_id}]\n{worker_agent_id} Task: {worker_task}&quot;)<br><br>            content_generator_worker_agents.append({<br>                &quot;agent_id&quot;: AgentId(worker_agent_id, f&quot;{self.id.key}/worker&quot;),<br>                &quot;task&quot;: ContentTask(agent_id=worker_agent_id, task=worker_task)<br>                }<br>            )<br>    <br>        content_generator_worker_agents_responses = await asyncio.gather(*[self.send_message(worker[&quot;task&quot;], worker[&quot;agent_id&quot;]) for worker in content_generator_worker_agents])<br>        <br>        for item in content_generator_worker_agents_responses:<br>            print(f&quot;\n[{self._agent_id}]\n{item.agent_id} Response:\n{item.text}&quot;)<br><br><br>        # File Store Worker Agent<br>        print(f&quot;{&#39;-&#39;*80}\n[{self._agent_id}]\nDelete Files&quot;)<br>        <br>        file_store_worker_agent_id = AgentId(FILE_STORE_WORKER_AGENT_ID, f&quot;{self.id.key}/worker&quot;)<br>        file_store_worker_agent_task = FileStoreTask(action_type=FileActionType.DELETE.value, file_store_metadata=file_store_metadata)<br>        <br>        print(f&quot;\n[{self._agent_id}]\n{FILE_STORE_WORKER_AGENT_ID} Task: Delete Files&quot;)<br><br>        file_store_worker_agent_response = await self.send_message(file_store_worker_agent_task, file_store_worker_agent_id)<br>        file_store_metadata = file_store_worker_agent_response.file_store_metadata<br><br>        print(f&quot;\n[{self._agent_id}]\n{FILE_STORE_WORKER_AGENT_ID} Response:\nFiles Deleted: {file_store_metadata}&quot;)<br><br><br>        # Final Result<br>        print(f&quot;{&#39;-&#39;*80}\n[{self._agent_id}]\nFinal Result&quot;)<br>        system_prompt = self._prompt<br>        system_prompt += &quot;\n\n## Content:\n&quot; + &quot;\n&quot;.join([response.text for response in content_generator_worker_agents_responses])<br><br>        model_result = await self._model_client.create(<br>            [SystemMessage(content=system_prompt), UserMessage(content=message.task, source=&quot;user&quot;)]<br>        )<br>        return FinalResult(result=model_result.content)<br><br><br><br>class FileStoreWorkerAgent(RoutedAgent):<br>    def __init__(<br>        self,<br>        open_ai_client: AsyncOpenAI,<br>        agent_id: str<br>    ) -&gt; None:<br>        super().__init__(description=agent_id)<br>        self._open_ai_client = open_ai_client<br>        self._agent_id = agent_id<br><br>    @message_handler<br>    async def handle_message(self, message: FileStoreTask, ctx: MessageContext) -&gt; FileStoreResponse:<br>        file_store_metadata = message.file_store_metadata<br><br>        if message.action_type == FileActionType.UPLOAD.value:<br>            file_store_metadata = await upload_files(self._open_ai_client, file_store_metadata)<br>        elif message.action_type == FileActionType.DELETE.value:<br>            file_store_metadata = await delete_files(self._open_ai_client, file_store_metadata)<br><br>        return FileStoreResponse(file_store_metadata=file_store_metadata)<br><br><br>class ContentGeneratorWorkerAgent(RoutedAgent):<br>    def __init__(<br>        self,<br>        open_ai_client: AsyncOpenAI,<br>        agent_id: str,<br>        prompt: str,<br>        file_store_metadata: FileStoreMetadata<br>    ) -&gt; None:<br>        super().__init__(description=agent_id)<br>        self._open_ai_client = open_ai_client<br>        self._agent_id = agent_id<br>        self._prompt = prompt<br>        self._file_store_metadata = file_store_metadata<br><br>    @message_handler<br>    async def handle_message(self, message: ContentTask, ctx: MessageContext) -&gt; ContentResponse:<br>        result = await generate_response(self._open_ai_client, self._prompt, message.task, self._file_store_metadata)<br>        return ContentResponse(agent_id=message.agent_id, text=result)<br><br><br><br>async def generate_response(open_ai_client: OpenAI, prompt: str, task: str, file_store_metadata: FileStoreMetadata):<br>    # Append task to prompt<br>    full_prompt = f&quot;{prompt}\n## Instructions:{task}&quot; if task else prompt<br><br>    if USE_VECTOR_STORE:<br>        # Use file search tool in request<br>        vector_store_id = file_store_metadata[&quot;vector_store_id&quot;]<br>        tools = [{&quot;type&quot;: &quot;file_search&quot;, &quot;vector_store_ids&quot;: [vector_store_id]}]<br>        input_data = full_prompt<br>    else:<br>        # Use file inputs in request<br>        file_ids = file_store_metadata[&quot;file_ids&quot;]<br>        tools = None<br>        input_data = [<br>            {<br>                &quot;role&quot;: &quot;user&quot;,<br>                &quot;content&quot;: [<br>                    {&quot;type&quot;: &quot;input_file&quot;, &quot;file_id&quot;: file_id} for file_id in file_ids<br>                ] + [<br>                    {&quot;type&quot;: &quot;input_text&quot;, &quot;text&quot;: full_prompt}<br>                ]<br>            }<br>        ]<br><br>    # Call OpenAI Responses API<br>    response = await open_ai_client.responses.create(<br>        model=OPEN_AI_MODEL,<br>        input=input_data,<br>        tool_choice=&quot;required&quot; if USE_VECTOR_STORE else None,<br>        tools=tools<br>    )<br><br>    # Return response text<br>    return response.output_text<br><br><br><br>async def upload_files(open_ai_client: OpenAI, file_store_metadata: any):<br>    if USE_VECTOR_STORE:<br>        # Create vector store and add uploaded files to vector store<br>        vs_id = await create_vector_store(open_ai_client, VECTOR_STORE_NAME, file_store_metadata)<br>        file_store_metadata[&quot;vector_store_id&quot;] = vs_id<br>    else:<br>        # Upload files<br>        file_paths = file_store_metadata[&quot;file_paths&quot;]<br>        tasks = [upload_file(open_ai_client, file_path) for file_path in file_paths]<br>        file_responses = await asyncio.gather(*tasks)<br><br>        file_ids = []<br>        for file in file_responses:<br>            file_ids.append(file.id)<br><br>        print(f&quot;\nFile IDs: {file_ids}&quot;)<br>        file_store_metadata[&quot;file_ids&quot;] = file_ids<br><br>    return file_store_metadata<br><br>async def delete_files(open_ai_client: OpenAI, file_store_metadata: any):<br>    if USE_VECTOR_STORE:<br>        # Delete vector store and files<br>        vs_id = file_store_metadata[&quot;vector_store_id&quot;]<br>        await delete_vector_store(open_ai_client, vs_id)<br>    else:<br>        # Delete files<br>        file_ids = file_store_metadata[&quot;file_ids&quot;]<br>        for file_id in file_ids:<br>            await delete_file(open_ai_client, file_id)<br><br>    return file_store_metadata<br><br><br>async def upload_file(open_ai_client: OpenAI, file_path: str):<br>    # Upload file<br>    return await open_ai_client.files.create(<br>        file=get_file(file_path),<br>        purpose=&quot;user_data&quot;<br>    )<br><br>def get_file(file_path):<br>    if file_path.startswith(&quot;http://&quot;) or file_path.startswith(&quot;https://&quot;):<br>        # Get remote file<br>        response = requests.get(file_path)<br>        file_content = BytesIO(response.content)<br>        file_name = file_path.split(&quot;/&quot;)[-1]<br>        file_object = (file_name, file_content)<br>        return file_object<br>    else:<br>        # Get local file<br>        file_object = open(file_path, &quot;rb&quot;)<br>        return file_object<br><br><br>async def delete_file(open_ai_client: OpenAI, file_id: str):<br>    # Delete file by file ID<br>    await open_ai_client.files.delete(file_id)<br><br><br>async def create_vector_store(open_ai_client: OpenAI, vector_store_name: str, docs_metadata: any):<br>    file_paths = docs_metadata[&quot;file_paths&quot;]<br>    print(f&quot;\nVector Store File Paths: {file_paths}&quot;)<br>    <br>    # Create vector store<br>    vector_store = await open_ai_client.vector_stores.create(name=vector_store_name)<br>    vector_store_id = vector_store.id<br>    print(vector_store_id)<br><br>    # Ready the files for upload to OpenAI<br>    file_streams = [get_file(path) for path in file_paths]<br>    print(file_streams)<br><br>    # Use the upload and poll SDK helper to upload the files, add them to the vector store,<br>    # and poll the status of the file batch for completion.<br>    file_batch = await open_ai_client.vector_stores.file_batches.upload_and_poll(vector_store_id=vector_store_id, files=file_streams)<br><br>    # Print the status and the file counts of the batch to see the result of this operation.<br>    print(file_batch.status)<br>    print(file_batch.file_counts)<br><br>    # Get all files in vector store<br>    result = open_ai_client.vector_stores.files.list(vector_store_id=vector_store_id)<br>    print(result)<br><br>    return vector_store_id<br><br><br>async def delete_vector_store(open_ai_client: OpenAI, vector_store_id: str):<br>    # Get all files in vector store<br>    vector_store_files = await open_ai_client.vector_stores.files.list(<br>        vector_store_id=vector_store_id<br>    )<br><br>    # Delete files<br>    for file in vector_store_files.data:<br>        await open_ai_client.files.delete(file.id)<br><br>    # Delete vector store<br>    deleted_vector_store = await open_ai_client.vector_stores.delete(vector_store_id=vector_store_id)<br>    print(deleted_vector_store)<br><br>    return vector_store_id<br><br><br>async def execute():<br>    # Set input files<br>    file_store_metadata = {<br>       &quot;file_paths&quot;: [&quot;sample-doc1.pdf&quot;, &quot;sample-doc2.pdf&quot;, &quot;sample-doc3.pdf&quot;]<br>    }<br>    <br>    # Create OpenAI client<br>    open_ai_client = AsyncOpenAI(api_key=OPEN_AI_API_KEY, timeout=120.0, max_retries=3)<br>    runtime = SingleThreadedAgentRuntime()<br><br>    await SupervisorAgent.register(<br>        runtime,<br>        SUPERVISOR_AGENT_ID,<br>        lambda: SupervisorAgent(<br>            runtime=runtime, model_client=OpenAIChatCompletionClient(model=OPEN_AI_MODEL, api_key=OPEN_AI_API_KEY), open_ai_client=open_ai_client, agent_id=SUPERVISOR_AGENT_ID, prompt=SUPERVISOR_AGENT_PROMPT, file_store_metadata=file_store_metadata<br>        ),<br>    )<br><br>    agent = AgentId(SUPERVISOR_AGENT_ID, &quot;default&quot;)<br><br>    runtime.start()<br>    final_result = await runtime.send_message(UserTask(task=USER_TASK), agent)<br>    response = json.dumps(final_result.__dict__)<br>    <br>    print(f&quot;\n[{response}]&quot;)<br>    <br>    await runtime.stop_when_idle()<br><br><br>asyncio.run(execute())</pre><h4>Run</h4><p>Run the Python script <strong>sample.py</strong> in terminal:</p><pre>python sample.py</pre><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*7J1-o4Hp6lUKRDttEU2R5Q.png" /><figcaption>Sample Output</figcaption></figure><h3>References</h3><ul><li><a href="https://microsoft.github.io/autogen/stable/">AutoGen</a></li><li><a href="https://platform.openai.com/docs/overview">OpenAI Developer Platform</a></li><li><a href="https://platform.openai.com/docs/api-reference/introduction">OpenAI API Reference</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=ce395c4929db" width="1" height="1" alt=""><hr><p><a href="https://medium.com/oracle-saas-paas/autogen-multi-agent-system-integrated-with-openai-file-search-ce395c4929db">AutoGen — Multi-Agent System Integrated with OpenAI File Search</a> was originally published in <a href="https://medium.com/oracle-saas-paas">oracle-saas-paas</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AutoGen — Orchestrator-Worker Agents Design Pattern]]></title>
            <link>https://medium.com/oracle-saas-paas/autogen-orchestrator-worker-agents-design-pattern-eef8698459b2?source=rss-f49c260596a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/eef8698459b2</guid>
            <category><![CDATA[ai-agent]]></category>
            <category><![CDATA[openai]]></category>
            <category><![CDATA[autogen]]></category>
            <category><![CDATA[agentic-system]]></category>
            <category><![CDATA[oracle-cloud]]></category>
            <dc:creator><![CDATA[Steve Zebib]]></dc:creator>
            <pubDate>Mon, 14 Apr 2025 14:23:03 GMT</pubDate>
            <atom:updated>2025-07-25T14:09:55.852Z</atom:updated>
            <content:encoded><![CDATA[<h3>AutoGen — Orchestrator-Worker Agents Design Pattern</h3><h3>Overview</h3><p>This article demonstrates how to implement AutoGen multi-agent design pattern using orchestrator and worker agents.</p><p>The orchestrator agent takes user input and dispatches tasks to multiple worker agents. Each worker agent has specific skills required to independently accomplish an assigned task. The orchestrator agent aggregates all the results from the worker agents and returns a final result.</p><h3>Setup</h3><h4>OpenAI</h4><ul><li><a href="https://platform.openai.com/api-keys">OpenAI API Key</a></li></ul><h4>Environment</h4><p>The following configuration is used in this example:</p><ul><li><a href="https://docs.anaconda.com/miniconda/">Conda for Mac</a></li><li>Python 3.10</li></ul><p>The following is a script to setup the env and install packages:</p><pre>conda create -n autogenDev python=3.10    <br>conda activate autogenDev<br>pip install &#39;autogen-agentchat==0.4.0.dev11&#39;<br>pip install &#39;autogen-ext[openai]==0.4.0.dev11&#39;</pre><h3>Sample</h3><p>In this example, we’ll develop an agentic system that utilizes orchestrator and worker agents to build a task management solution.</p><p>User Prompt:</p><blockquote>“I want to create a system to manage tasks. The system should allow creating tasks, assigning them to team members, updating their status, and generating a summary of pending tasks.”</blockquote><p>Agents:</p><ul><li>Orchestrator Agent (Software Director): Orchestrator Agent acting as a software director with deep technical expertise to lead software development, manage teams effectively, and deliver innovative, high-quality products that align seamlessly with business objectives.</li><li>Worker Agent (Product Manager): Worker Agent acting as a product manager who drives the product vision, roadmap, and feature prioritization, ensuring alignment with user needs and business goals through stakeholder collaboration and timely, budget-conscious execution.</li><li>Worker Agent (Software Developer): Worker Agent acting as a software developer with strong programming skills, problem-solving abilities, and attention to detail to design, build, and maintain efficient, scalable, and reliable software solutions.</li><li>Worker Agent (Business Analyst): Worker Agent acting as a business analyst skilled in understanding business needs and technical solutions through meticulous analysis, clear requirement gathering, and effective stakeholder communication to ensure projects align with strategic goals and deliver measurable value.</li></ul><h4>Code</h4><p>Create a Python script <strong>sample.py</strong> and copy/paste the following:</p><pre>import asyncio<br>from dataclasses import dataclass<br>from typing import List<br><br>from autogen_core import AgentId, MessageContext, RoutedAgent, SingleThreadedAgentRuntime, message_handler<br>from autogen_core.models import ChatCompletionClient, SystemMessage, UserMessage<br>from autogen_ext.models.openai import OpenAIChatCompletionClient<br><br>OPEN_AI_API_KEY = &quot;&lt;OPEN_AI_KEY&gt;&quot;<br>OPEN_AI_MODEL = &quot;gpt-4o&quot;<br><br>ORCHESTRATOR_AGENT_ID = &quot;Orchestrator_Agent&quot;<br>PRODUCT_MANAGER_AGENT_ID = &quot;Product_Manager_Agent&quot;<br>SOFTWARE_DEVELOPER_AGENT_ID = &quot;Software_Developer_Agent&quot;<br>BUSINESS_ANALYST_AGENT_ID = &quot;Business_Analyst_Agent&quot;<br><br>USER_TASK = &quot;I want to create a system to manage tasks. The system should allow creating tasks, assigning them to team members, updating their status, and generating a summary of pending tasks.&quot;<br>ORCHESTRATOR_AGENT_PROMPT = &quot;You are an AI assistant thinking as a software director with deep technical expertise to lead software development, manage teams effectively, and deliver innovative, high-quality products that align seamlessly with business objectives. Analyze all responses provided by the worker agents and provide comprehensive report which includes: requirements, objectives, and sample code.&quot;<br>PRODUCT_MANAGER_AGENT_PROMPT = &quot;You are an AI assistant thinking as a product manager who drives the product vision, roadmap, and feature prioritization, ensuring alignment with user needs and business goals through stakeholder collaboration and timely, budget-conscious execution. Define objectives and requirements required to develop this system.&quot;<br>SOFTWARE_DEVELOPER_AGENT_PROMPT = &quot;You are an AI assistant thinking as a software developer with strong programming skills, problem-solving abilities, and attention to detail to design, build, and maintain efficient, scalable, and reliable software solutions. Write sample code using JavaScript to develop this system based on specific requirements.&quot;<br>BUSINESS_ANALYST_AGENT_PROMPT = &quot;You are an AI assistant thinking as a business analyst skilled in understanding business needs and technical solutions through meticulous analysis, clear requirement gathering, and effective stakeholder communication to ensure projects align with strategic goals and deliver measurable value. Analyze the objectives and requirements and provide constructive feedback.&quot;<br><br>@dataclass<br>class WorkerTask:<br>    task: str<br>    previous_results: List[str]<br><br><br>@dataclass<br>class WorkerTaskResult:<br>    result: str<br><br><br>@dataclass<br>class UserTask:<br>    task: str<br><br><br>@dataclass<br>class FinalResult:<br>    result: str<br><br><br>class WorkerAgent(RoutedAgent):<br>    def __init__(<br>        self,<br>        model_client: ChatCompletionClient,<br>        agent_id: str,<br>        prompt: str<br>    ) -&gt; None:<br>        super().__init__(description=agent_id)<br>        self._agent_id = agent_id<br>        self._model_client = model_client<br>        self._prompt = prompt<br><br>    @message_handler<br>    async def handle_task(self, message: WorkerTask, ctx: MessageContext) -&gt; WorkerTaskResult:<br>        system_prompt = self._prompt<br>        if message.previous_results:<br>            # If previous results are provided, we need to synthesize them to create a single prompt.<br>            system_prompt += &quot;\n&quot; + &quot;\n\n&quot;.join([f&quot;{i+1}. {r}&quot; for i, r in enumerate(message.previous_results)])<br>            model_result = await self._model_client.create(<br>                [SystemMessage(content=system_prompt), UserMessage(content=message.task, source=&quot;user&quot;)]<br>            )<br>        else:<br>            # If no previous results are provided, we can simply pass the user query to the model.<br>            model_result = await self._model_client.create(<br>                [SystemMessage(content=system_prompt), UserMessage(content=message.task, source=&quot;user&quot;)])<br>        assert isinstance(model_result.content, str)<br>        print(f&quot;{&#39;-&#39;*80}\nWorker-{self.id}:\n{model_result.content}&quot;)<br>        return WorkerTaskResult(result=model_result.content)<br><br>class OrchestratorAgent(RoutedAgent):<br>    def __init__(<br>        self,<br>        model_client: ChatCompletionClient,<br>        agent_id: str,<br>        prompt: str,<br>        worker_agent_types: List[str],<br>        num_layers: int,<br>    ) -&gt; None:<br>        super().__init__(description=agent_id)<br>        self._agent_id = agent_id<br>        self._model_client = model_client<br>        self._prompt = prompt<br>        self._worker_agent_types = worker_agent_types<br>        self._num_layers = num_layers<br><br>    @message_handler<br>    async def handle_task(self, message: UserTask, ctx: MessageContext) -&gt; FinalResult:<br>        print(f&quot;{&#39;-&#39;*80}\nOrchestrator-{self.id}:\nReceived task: {message.task}&quot;)<br>        # Create task for the first layer.<br>        worker_task = WorkerTask(task=message.task, previous_results=[])<br>        # Iterate over layers.<br>        for i in range(self._num_layers - 1):<br>            # Assign workers for this layer.<br>            worker_ids = [<br>                AgentId(worker_type, f&quot;{self.id.key}/layer_{i}/worker_{j}&quot;)<br>                for j, worker_type in enumerate(self._worker_agent_types)<br>            ]<br>            # Dispatch tasks to workers.<br>            print(f&quot;{&#39;-&#39;*80}\nOrchestrator-{self.id}:\nDispatch to workers at layer {i}&quot;)<br>            results = await asyncio.gather(*[self.send_message(worker_task, worker_id) for worker_id in worker_ids])<br>            print(f&quot;{&#39;-&#39;*80}\nOrchestrator-{self.id}:\nReceived results from workers at layer {i}&quot;)<br>            # Prepare task for the next layer.<br>            worker_task = WorkerTask(task=message.task, previous_results=[r.result for r in results])<br>        # Perform final aggregation.<br>        print(f&quot;{&#39;-&#39;*80}\nOrchestrator-{self.id}:\nPerforming final aggregation&quot;)<br>        system_prompt = self._prompt<br>        system_prompt += &quot;\n&quot; + &quot;\n\n&quot;.join([f&quot;{i+1}. {r}&quot; for i, r in enumerate(worker_task.previous_results)])<br>        model_result = await self._model_client.create(<br>            [SystemMessage(content=system_prompt), UserMessage(content=message.task, source=&quot;user&quot;)]<br>        )<br>        assert isinstance(model_result.content, str)<br>        return FinalResult(result=model_result.content)<br><br><br>async def execute():<br>    runtime = SingleThreadedAgentRuntime()<br><br>    await WorkerAgent.register(<br>        runtime, PRODUCT_MANAGER_AGENT_ID, lambda: WorkerAgent(model_client=OpenAIChatCompletionClient(model=OPEN_AI_MODEL, api_key=OPEN_AI_API_KEY), agent_id=PRODUCT_MANAGER_AGENT_ID, prompt=PRODUCT_MANAGER_AGENT_PROMPT)<br>    )<br><br>    await WorkerAgent.register(<br>        runtime, SOFTWARE_DEVELOPER_AGENT_ID, lambda: WorkerAgent(model_client=OpenAIChatCompletionClient(model=OPEN_AI_MODEL, api_key=OPEN_AI_API_KEY), agent_id=SOFTWARE_DEVELOPER_AGENT_ID, prompt=SOFTWARE_DEVELOPER_AGENT_PROMPT)<br>    )<br><br>    await WorkerAgent.register(<br>        runtime, BUSINESS_ANALYST_AGENT_ID, lambda: WorkerAgent(model_client=OpenAIChatCompletionClient(model=OPEN_AI_MODEL, api_key=OPEN_AI_API_KEY), agent_id=BUSINESS_ANALYST_AGENT_ID, prompt=BUSINESS_ANALYST_AGENT_PROMPT)<br>    )<br><br>    await OrchestratorAgent.register(<br>        runtime,<br>        ORCHESTRATOR_AGENT_ID,<br>        lambda: OrchestratorAgent(<br>            model_client=OpenAIChatCompletionClient(model=OPEN_AI_MODEL, api_key=OPEN_AI_API_KEY), agent_id=ORCHESTRATOR_AGENT_ID, prompt=ORCHESTRATOR_AGENT_PROMPT, worker_agent_types=[PRODUCT_MANAGER_AGENT_ID, SOFTWARE_DEVELOPER_AGENT_ID, BUSINESS_ANALYST_AGENT_ID], num_layers=2<br>        ),<br>    )<br><br>    runtime.start()<br>    result = await runtime.send_message(UserTask(task=USER_TASK), AgentId(ORCHESTRATOR_AGENT_ID, &quot;default&quot;))<br>    await runtime.stop_when_idle()<br>    print(f&quot;{&#39;-&#39;*80}\nFinal result:\n{result.result}&quot;)<br><br><br>asyncio.run(execute())</pre><h4>Run</h4><p>Run the Python script <strong>sample.py</strong> in terminal:</p><pre>python sample.py</pre><p>The following is the final result returned by the Orchestrator Agent:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*u0kq4XC_XAAoQ9RT6lcx_w.png" /><figcaption>Sample Output</figcaption></figure><h3>References</h3><ul><li><a href="https://microsoft.github.io/autogen/stable/">AutoGen</a></li><li><a href="https://platform.openai.com/docs/api-reference/introduction">OpenAI API Reference</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=eef8698459b2" width="1" height="1" alt=""><hr><p><a href="https://medium.com/oracle-saas-paas/autogen-orchestrator-worker-agents-design-pattern-eef8698459b2">AutoGen — Orchestrator-Worker Agents Design Pattern</a> was originally published in <a href="https://medium.com/oracle-saas-paas">oracle-saas-paas</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[AutoGen — Cohere Integration]]></title>
            <link>https://medium.com/oracle-saas-paas/autogen-with-cohere-31c561be81f8?source=rss-f49c260596a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/31c561be81f8</guid>
            <category><![CDATA[gen-ai-tools]]></category>
            <category><![CDATA[autogen]]></category>
            <category><![CDATA[cohere]]></category>
            <category><![CDATA[oracle-cloud]]></category>
            <dc:creator><![CDATA[Steve Zebib]]></dc:creator>
            <pubDate>Sat, 14 Dec 2024 00:46:45 GMT</pubDate>
            <atom:updated>2025-07-24T21:22:27.080Z</atom:updated>
            <content:encoded><![CDATA[<h3>AutoGen — Cohere Integration</h3><h3>Overview</h3><p>This article explains how to use AutoGen with Cohere, an alternative to the default OpenAI integration. This includes setup instructions and an example to showcase AutoGen’s features with Cohere.</p><h3>Setup</h3><h4><strong>Cohere</strong></h4><ul><li>API Key from <a href="https://cohere.com/">Cohere</a></li></ul><h4>Environment</h4><p>The following configuration is used in this example:</p><ul><li><a href="https://docs.anaconda.com/miniconda/">Conda for Mac</a></li><li>Python 3.10 (compatible with AutoGen Cohere library 0.2.39)</li></ul><p>The following is a script to setup the env and install packages:</p><pre>conda create -n autogenDev python=3.10    <br>conda activate autogenDev<br>pip install &#39;autogen-agentchat[cohere]==0.2.39&#39;</pre><h3>Sample</h3><p>In this example, we’ll utilize the group chat multi-agent design pattern. We will create a group chat with the following:</p><p>User Prompt: “I want to learn more about AutoGen”</p><p>Agents:</p><ul><li>User Proxy Agent: Agent who will initiate the chat using the user prompt.</li><li>Assistant Agent (Basic): Agent with basic knowledge about AutoGen.</li><li>Assistant Agent (Professional): Agent with advanced knowledge about AutoGen and potential use cases.</li></ul><h4>Code</h4><p>Create a Python script (e.g. sample.py) and copy/paste the following:</p><pre>from autogen import AssistantAgent, UserProxyAgent, GroupChat, GroupChatManager<br><br># Prompts<br>userPrompt=&quot;I want to learn more about AutoGen.&quot;<br>basicAssistantAgentPrompt=&quot;You are a helpful AI assistant with basic knowledge of AutoGen.&quot;<br>professionalAssistantAgentPrompt=&quot;You are a helpful AI assistant with advanced knowledge of AutoGen and its use cases. Analyze response by Basic_Assistant_Agent and provide more information.&quot;<br><br># Cohere config<br>config_list = [<br>    {<br>        &quot;model&quot;: &quot;command-r-plus-08-2024&quot;,<br>        &quot;api_key&quot;: &quot;&lt;COHERE_API_KEY&gt;&quot;,<br>        &quot;api_type&quot;: &quot;cohere&quot;,<br>    }<br>]<br><br># Create user proxy agent<br>user_proxy_agent = UserProxyAgent(<br>    name=&quot;User_Proxy_Agent&quot;,<br>    llm_config={&quot;config_list&quot;: config_list},<br>    code_execution_config=False,<br>    human_input_mode=&quot;NEVER&quot;,<br>    max_consecutive_auto_reply=1,<br>)<br><br># Create the basic assistant agent<br>basic_assistant_agent = AssistantAgent(<br>    name=&quot;Basic_Assistant_Agent&quot;,<br>    llm_config={&quot;config_list&quot;: config_list},<br>    system_message=basicAssistantAgentPrompt,<br>    max_consecutive_auto_reply=1<br>)<br><br># Create the professional assistant agent<br>professional_assistant_agent = AssistantAgent(<br>    name=&quot;Professional_Assistant_Agent&quot;,<br>    llm_config={&quot;config_list&quot;: config_list},<br>    system_message=professionalAssistantAgentPrompt,<br>    max_consecutive_auto_reply=1,<br>)<br><br># Create a group chat between all agents<br>groupChat = GroupChat(<br>    agents=[user_proxy_agent, basic_assistant_agent, professional_assistant_agent],<br>    messages=[],<br>    max_round=3,<br>)<br><br>groupChatManager = GroupChatManager(<br>    groupchat=groupChat,<br>    llm_config={&quot;config_list&quot;: config_list}<br>)<br><br># Initiate the chat using the User Proxy Agent and pass the user prompt<br>chat_result = user_proxy_agent.initiate_chat(<br>    groupChatManager,<br>    message=userPrompt<br>)</pre><h4>Run</h4><p>Run the Python script (e.g. sample.py) in terminal:</p><pre>python sample.py</pre><p>The following is the final output returned by both the Basic Assistant Agent and Professional Assistant Agent:</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*wU8bstJBp5dz1ttYA58h6g.png" /><figcaption>Demo Output</figcaption></figure><h3>References</h3><ul><li><a href="https://microsoft.github.io/autogen/0.2/docs/topics/non-openai-models/cloud-cohere/">AutoGen — Cohere</a></li><li><a href="https://docs.cohere.com/reference/about">Cohere API Reference</a></li></ul><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=31c561be81f8" width="1" height="1" alt=""><hr><p><a href="https://medium.com/oracle-saas-paas/autogen-with-cohere-31c561be81f8">AutoGen — Cohere Integration</a> was originally published in <a href="https://medium.com/oracle-saas-paas">oracle-saas-paas</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[VBS: Opportunities Layout — Custom Sales Stage Popup]]></title>
            <link>https://medium.com/oracle-jet/vbs-opportunities-layout-custom-sales-stage-popup-8c98bc2e564f?source=rss-f49c260596a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/8c98bc2e564f</guid>
            <category><![CDATA[oracle-cloud]]></category>
            <category><![CDATA[oracle-visual-builder]]></category>
            <dc:creator><![CDATA[Steve Zebib]]></dc:creator>
            <pubDate>Tue, 03 Dec 2024 01:43:23 GMT</pubDate>
            <atom:updated>2024-12-13T18:10:51.153Z</atom:updated>
            <content:encoded><![CDATA[<h3>VBS: Opportunities Layout — Custom Sales Stage Popup</h3><p><strong><em>DISCLAIMER: </em></strong><em>The views expressed in this story are my own and do not necessarily reflect the views of Oracle.</em></p><h3>Overview</h3><p>This article describes how to create and enable a custom sales stage popup on the opportunities details page in Visual Builder Studio. The setup includes the following:</p><ul><li>Fragment: Create sales stage popup</li><li>Opportunities Details Page: Add the fragment</li><li>Opportunities Header Layout: Add an event to open the fragment when sales stage badge is clicked</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*I1RPOcJ551y4HxslIU35OA.png" /></figure><h3>Setup</h3><h4>Fragment (Sales Stage Popup)</h4><ul><li>Create a fragment called “sales-stage-popup”.</li><li>The sales stage popup will show all sales stages and status for each stage (CURRENT, PENDING, COMPLETED).</li><li>Select current sales stage by clicking on sales stage.</li><li>NOTE: The input parameters for the fragment which need to be passed from from the container (opportunities layout) include: optyNumber, currentStgId, and salesStagesList.</li><li>In the fragment, copy all the code shown below including: Fragment JSON, HTML, JS, and Action Chains (JS).</li></ul><p><strong>JSON</strong></p><pre>{<br>  &quot;description&quot;: &quot;&quot;,<br>  &quot;eventListeners&quot;: {<br>    &quot;divClick&quot;: {<br>      &quot;chains&quot;: [<br>        {<br>          &quot;chain&quot;: &quot;updateSalesStageActionChain&quot;,<br>          &quot;parameters&quot;: {<br>            &quot;stgId&quot;: &quot;[[ $current.data.StgId ]]&quot;<br>          }<br>        }<br>      ]<br>    },<br>    &quot;divClick2&quot;: {<br>      &quot;chains&quot;: [<br>        {<br>          &quot;chain&quot;: &quot;updateSalesStageActionChain&quot;,<br>          &quot;parameters&quot;: {<br>            &quot;stgId&quot;: &quot;[[ $current.data.StgId ]]&quot;<br>          }<br>        }<br>      ]<br>    },<br>    &quot;divClick3&quot;: {<br>      &quot;chains&quot;: [<br>        {<br>          &quot;chain&quot;: &quot;updateSalesStageActionChain&quot;,<br>          &quot;parameters&quot;: {<br>            &quot;stgId&quot;: &quot;[[ $current.data.StgId ]]&quot;<br>          }<br>        }<br>      ]<br>    },<br>    &quot;vbEnter&quot;: {<br>      &quot;chains&quot;: [<br>        {<br>          &quot;chain&quot;: &quot;updateSalesStageActionChain&quot;,<br>          &quot;parameters&quot;: {<br>            &quot;stgId&quot;: &quot;{{ $variables.currentStgId }}&quot;<br>      }<br>        }<br>      ]<br>    }<br>  },<br>  &quot;events&quot;: {<br>    &quot;updateSalesStage&quot;: {<br>      &quot;propagationBehavior&quot;: &quot;container&quot;,<br>      &quot;description&quot;: &quot;Emits to container&quot;,<br>      &quot;payloadType&quot;: {<br>        &quot;OptyNumber&quot;: &quot;string&quot;,<br>        &quot;StgId&quot;: &quot;number&quot;<br>      }<br>    }<br>  },<br>  &quot;imports&quot;: {<br>    &quot;components&quot;: {<br>      &quot;oj-popup&quot;: {<br>        &quot;path&quot;: &quot;ojs/ojpopup&quot;<br>      }<br>    }<br>  },<br>  &quot;types&quot;: {<br>    &quot;salesStageType&quot;: {<br>      &quot;Name&quot;: &quot;string&quot;,<br>      &quot;StgId&quot;: &quot;number&quot;,<br>      &quot;StgOrder&quot;: &quot;number&quot;<br>    }<br>  },<br>  &quot;variables&quot;: {<br>    &quot;currentSalesStage&quot;: {<br>      &quot;type&quot;: &quot;salesStageType&quot;,<br>      &quot;required&quot;: false,<br>      &quot;input&quot;: &quot;none&quot;<br>    },<br>    &quot;currentStgId&quot;: {<br>      &quot;type&quot;: &quot;number&quot;,<br>      &quot;required&quot;: true,<br>      &quot;input&quot;: &quot;fromCaller&quot;<br>    },<br>    &quot;dataReady&quot;: {<br>      &quot;type&quot;: &quot;boolean&quot;,<br>      &quot;defaultValue&quot;: false<br>    },<br>    &quot;optyNumber&quot;: {<br>      &quot;type&quot;: &quot;string&quot;,<br>      &quot;required&quot;: true,<br>      &quot;input&quot;: &quot;fromCaller&quot;<br>    },<br>    &quot;salesStagesList&quot;: {<br>      &quot;type&quot;: &quot;salesStageType[]&quot;,<br>      &quot;required&quot;: true,<br>      &quot;input&quot;: &quot;fromCaller&quot;<br><br>    },<br>    &quot;salesStageStatusList&quot;: {<br>      &quot;required&quot;: false,<br>      &quot;type&quot;: &quot;any[]&quot;,<br>      &quot;input&quot;: &quot;none&quot;<br>    },<br>    &quot;salesStageStatusListADP&quot;: {<br>      &quot;type&quot;: &quot;vb/ArrayDataProvider2&quot;,<br>      &quot;defaultValue&quot;: {<br>        &quot;data&quot;: &quot;{{ $variables.salesStageStatusList }}&quot;,<br>        &quot;itemType&quot;: &quot;any&quot;<br>      }<br>    }<br><br>  }<br>}</pre><p><strong>HTML</strong></p><pre>&lt;style&gt;<br>  .sales-stage {<br>    width: 170px;<br>    height: 150px;<br>    font-size: 14px;<br>    text-align: center;<br>    color: #FFF;<br>    margin: 4px;<br>    padding: 4px 4px 4px 16px;<br>    float: left;<br>    position: relative;<br>    background-color: var(--oj-core-neutral-1);<br>    cursor: pointer;<br>  }<br><br>  .sales-stage:after,<br>  .sales-stage:before {<br>    content: &quot; &quot;;<br>    position: absolute;<br>    top: 0;<br>    right: -16px;<br>    width: 0;<br>    height: 0;<br>    border-top: 76px solid transparent;<br>    border-bottom: 82px solid transparent;<br>    border-left: 17px solid var(--oj-core-neutral-1);<br>    z-index: 2;<br>  }<br><br>  .sales-stage:before {<br>    right: auto;<br>    left: 0;<br>    border-left: 16px solid #fff;<br>    z-index: 0;<br>  }<br><br>  .sales-stage:first-child:before,<br>  .sales-stage:last-child:after {<br>    border: none;<br>  }<br><br>  .sales-stage:first-child {<br>    border-top-left-radius: 24px;<br>    border-bottom-left-radius: 24px;<br>  }<br><br>  .sales-stage:last-child {<br>    border-top-right-radius: 24px;<br>    border-bottom-right-radius: 24px;<br>  }<br><br>  .sales-stage.current {<br>    color: #fff;<br>    background-color: var(--oj-core-info-1);<br>  }<br><br>  .sales-stage.current:after {<br>    border-left: 17px solid var(--oj-core-info-1);<br>  }<br><br>  .sales-stage.completed {<br>    color: #fff;<br>    background-color: var(--oj-core-success-1);<br>  }<br><br>  .sales-stage.completed:after {<br>    border-left: 17px solid var(--oj-core-success-1);<br>  }<br>&lt;/style&gt;<br><br>&lt;oj-popup id=&quot;popup-sales-stage&quot;&gt;<br>  &lt;div class=&quot;oj-flex oj-sm-align-items-center oj-sm-margin-6x&quot;&gt;<br>    &lt;oj-bind-if test=&quot;[[ $variables.dataReady ]]&quot;&gt;<br>      &lt;oj-bind-for-each data=&quot;[[ $variables.salesStageStatusListADP ]]&quot;&gt;<br>        &lt;template&gt;<br>          &lt;oj-bind-if test=&#39;[[ $current.data.Status === &quot;COMPLETED&quot; ]]&#39;&gt;<br>            &lt;div class=&quot;sales-stage completed&quot; on-click=&quot;[[$listeners.divClick]]&quot;&gt;<br>              &lt;div<br>                class=&quot;oj-flex oj-sm-align-items-center oj-sm-justify-content-center oj-sm-padding-2x&quot; style=&quot;width: 100%; height: 100%;&quot;&gt;<br>                &lt;img :src=&quot;[[ $extension.path + $fragment.functions.getSalesStageImageUrl($current.data.StgOrder) ]]&quot; width=&quot;100&quot; height=&quot;100&quot; style=&quot;border-radius: 50%;&quot;&gt;<br>                &lt;span class=&quot;oj-sm-margin-1x-top&quot;&gt;&lt;oj-bind-text value=&quot;[[ $current.data.Name ]]&quot;&gt;&lt;/oj-bind-text&gt;&lt;/span&gt;<br>              &lt;/div&gt;<br>            &lt;/div&gt;<br>          &lt;/oj-bind-if&gt;<br>          &lt;oj-bind-if test=&#39;[[ $current.data.Status === &quot;CURRENT&quot; ]]&#39;&gt;<br>            &lt;div class=&quot;sales-stage current&quot; on-click=&quot;[[$listeners.divClick2]]&quot;&gt;<br>              &lt;div<br>                class=&quot;oj-flex oj-sm-align-items-center oj-sm-justify-content-center oj-sm-padding-2x&quot; style=&quot;width: 100%; height: 100%;&quot;&gt;<br>                &lt;img :src=&quot;[[ $extension.path + $fragment.functions.getSalesStageImageUrl($current.data.StgOrder) ]]&quot; width=&quot;100&quot; height=&quot;100&quot; style=&quot;border-radius: 50%;&quot;&gt;<br>                &lt;span class=&quot;oj-sm-margin-1x-top&quot;&gt;&lt;oj-bind-text value=&quot;[[ $current.data.Name ]]&quot;&gt;&lt;/oj-bind-text&gt;&lt;/span&gt;<br>              &lt;/div&gt;<br>            &lt;/div&gt;<br>          &lt;/oj-bind-if&gt;<br>          &lt;oj-bind-if test=&#39;[[ $current.data.Status === &quot;PENDING&quot; ]]&#39;&gt;<br>            &lt;div class=&quot;sales-stage&quot; on-click=&quot;[[$listeners.divClick3]]&quot;&gt;<br>              &lt;div<br>                class=&quot;oj-flex oj-sm-align-items-center oj-sm-justify-content-center oj-sm-padding-2x&quot; style=&quot;width: 100%; height: 100%;&quot;&gt;<br>                &lt;img :src=&quot;[[ $extension.path + $fragment.functions.getSalesStageImageUrl($current.data.StgOrder) ]]&quot; width=&quot;100&quot; height=&quot;100&quot; style=&quot;border-radius: 50%;&quot;&gt;<br>                &lt;span class=&quot;oj-sm-margin-1x-top&quot;&gt;&lt;oj-bind-text value=&quot;[[ $current.data.Name ]]&quot;&gt;&lt;/oj-bind-text&gt;&lt;/span&gt;<br>              &lt;/div&gt;<br>            &lt;/div&gt;<br>          &lt;/oj-bind-if&gt;<br>        &lt;/template&gt;<br>      &lt;/oj-bind-for-each&gt;<br>    &lt;/oj-bind-if&gt;<br>  &lt;/div&gt;<br>&lt;/oj-popup&gt;</pre><p><strong>JS</strong></p><pre>define([], () =&gt; {<br>  &#39;use strict&#39;;<br><br>  class FragmentModule {<br>  }<br><br>  FragmentModule.prototype.processCurrentSalesStage = function (salesStagesList, stgId) {<br>    let salesStagesListCopy = JSON.parse(JSON.stringify(salesStagesList));<br>    let currentSalesStage;<br><br>    for (let salesStage of salesStagesListCopy) {<br>      if (salesStage.StgId === stgId) {<br>        currentSalesStage = salesStage;<br>        break;<br>      }<br>    }<br><br>    return currentSalesStage;<br>  };<br><br>  FragmentModule.prototype.updateSalesStagesStatusList = function (salesStagesList, currentSalesStage) {<br>    let salesStagesListCopy = JSON.parse(JSON.stringify(salesStagesList));<br>    let currentSalesStageCopy = JSON.parse(JSON.stringify(currentSalesStage));<br>    <br>    for (let salesStage of salesStagesListCopy) {<br>      if (salesStage.StgOrder &lt; currentSalesStageCopy.StgOrder) {<br>        salesStage.Status = &quot;COMPLETED&quot;;<br>      } else if (salesStage.StgOrder &gt; currentSalesStageCopy.StgOrder) {<br>        salesStage.Status = &quot;PENDING&quot;;<br>      } else if (salesStage.StgOrder === currentSalesStageCopy.StgOrder) {<br>        salesStage.Status = &quot;CURRENT&quot;;<br>      }<br>    }<br><br>    return salesStagesListCopy;<br>  };<br><br>  // NOTE: Add sales stages images in resources directory. You can use a different image for each sales stage based on StgOrder value.<br>  FragmentModule.prototype.getSalesStageImageUrl = function (stgOrder) {<br>    return &quot;resources/images/o_tag.png&quot;;<br>  };<br><br>  return FragmentModule;<br>});</pre><p><strong>Action Chain (updateSalesStageActionChain)</strong></p><pre>define([<br>  &#39;vb/action/actionChain&#39;,<br>  &#39;vb/action/actions&#39;,<br>  &#39;vb/action/actionUtils&#39;,<br>], (<br>  ActionChain,<br>  Actions,<br>  ActionUtils<br>) =&gt; {<br>  &#39;use strict&#39;;<br><br>  class updateSalesStageActionChain extends ActionChain {<br><br>    /**<br>     * @param {Object} context<br>     * @param {Object} params<br>     * @param {number} params.stgId <br>     */<br>    async run(context, { stgId }) {<br>      const { $fragment, $extension, $global, $functions, $variables } = context;<br><br>      $variables.dataReady = false;<br><br>      const processCurrentSalesStage = await $functions.processCurrentSalesStage($variables.salesStagesList, stgId);<br><br>      $fragment.variables.currentSalesStage = processCurrentSalesStage;<br><br>      await Actions.fireEvent(context, {<br>        event: &#39;updateSalesStage&#39;,<br>        payload: {<br>          OptyNumber: $variables.optyNumber,<br>          StgId: stgId,<br>        },<br>      });<br><br>      const callFunctionResult = await $fragment.functions.updateSalesStagesStatusList($fragment.variables.salesStagesList, $fragment.variables.currentSalesStage);<br><br>      $fragment.variables.salesStageStatusList = callFunctionResult;<br><br>      await Actions.fireDataProviderEvent(context, {<br>        refresh: null,<br>        target: $fragment.variables.salesStageStatusListADP,<br>      });<br><br>      $fragment.variables.dataReady = true;<br>    }<br>  }<br><br>  return updateSalesStageActionChain;<br>});</pre><h4>Opportunities Details Page</h4><ul><li>Go to App UIs &gt; Sales &gt; opportunities &gt; opportunities-detail</li><li>Duplicate the default layout and a section (e.g. salesStagePopupTemplate) which will contain the fragment.</li><li>Add the fragment (sales-stage-popup) to the template as shown below.</li></ul><pre>&lt;template id=&quot;salesStagePopupTemplate&quot;&gt;<br>  &lt;oj-vb-fragment bridge=&quot;[[vbBridge]]&quot; name=&quot;sales-stage-popup&quot;&gt;<br>    &lt;oj-vb-fragment-param name=&quot;optyNumber&quot; value=&quot;[[ $base.variables.puid ]]&quot;&gt;&lt;/oj-vb-fragment-param&gt;<br>    &lt;oj-vb-fragment-param name=&quot;currentStgId&quot; value=&quot;[[ $variables.currentStgId ]]&quot;&gt;&lt;/oj-vb-fragment-param&gt;<br>    &lt;oj-vb-fragment-param name=&quot;salesStagesList&quot; value=&quot;[[ $variables.salesStagesList ]]&quot;&gt;&lt;/oj-vb-fragment-param&gt;<br>  &lt;/oj-vb-fragment&gt;<br>&lt;/template&gt;</pre><ul><li>Note the input parameters that need to be passed to the fragment include: optyNumber (puid), currentStgId (number), and salesStagesList (array of salesStageType). For example:</li></ul><pre>optyNumber = &quot;10001&quot;;<br>currentStgId = 300000000323740;<br>salesStagesList = [<br>    {<br>        &quot;Name&quot;: &quot;01 - Qualification&quot;,<br>        &quot;StgId&quot;: 300000000323741,<br>        &quot;StgOrder&quot;: 1<br>    },<br>    {<br>        &quot;Name&quot;: &quot;02 - Discovery&quot;,<br>        &quot;StgId&quot;: 300000000323740,<br>        &quot;StgOrder&quot;: 2<br>    },<br>    {<br>        &quot;Name&quot;: &quot;03 - Building Vision&quot;,<br>        &quot;StgId&quot;: 300000000323742,<br>        &quot;StgOrder&quot;: 3<br>    },<br>    {<br>        &quot;Name&quot;: &quot;04 - Presentation&quot;,<br>        &quot;StgId&quot;: 300000000323739,<br>        &quot;StgOrder&quot;: 4<br>    },<br>    {<br>        &quot;Name&quot;: &quot;05 - Agreement&quot;,<br>        &quot;StgId&quot;: 300000000323736,<br>        &quot;StgOrder&quot;: 5<br>    },<br>    {<br>        &quot;Name&quot;: &quot;06 - Negotiation&quot;,<br>        &quot;StgId&quot;: 300000000323738,<br>        &quot;StgOrder&quot;: 6<br>    },<br>    {<br>        &quot;Name&quot;: &quot;07 - Closed&quot;,<br>        &quot;StgId&quot;: 300000000323737,<br>        &quot;StgOrder&quot;: 7<br>    }<br>]<br></pre><h4>Opportunities Header Layout</h4><ul><li>Go to Layout &gt; Sales &gt; Opportunities &gt; Header Layout</li><li>Duplicate the default layout and make sure the SalesStage field is set as the badgeFieldName.</li><li>Add the opportunitiesOnLoadHeaderEvent which triggers an action chain (opportunitiesOnLoadHeaderEventListener) as shown below. This action chain will add a click handler on the SalesStage field to open the sales stage popup. Make sure the ID of popup (e.g. popup-sales-stage) is the same as the ID used in the fragment.</li></ul><p><strong>Action Chain (opportunitiesOnLoadHeaderEventListener)</strong></p><pre>define([<br>  &#39;vb/action/actionChain&#39;,<br>  &#39;vb/action/actions&#39;,<br>  &#39;vb/action/actionUtils&#39;,<br>], (<br>  ActionChain,<br>  Actions,<br>  ActionUtils<br>) =&gt; {<br>  &#39;use strict&#39;;<br><br>  class opportunitiesOnLoadHeaderEventListener extends ActionChain {<br><br>    /**<br>     * @param {Object} context<br>     */<br>    async run(context) {<br>      const { $layout, $base, $extension, $responsive, $user } = context;<br><br>      const elements = document.getElementsByClassName(&quot;cx-badge-item&quot;);<br><br>      if (elements &amp;&amp; elements.length &gt; 0) {<br>        let element = elements.item(0);<br>        element.addEventListener(&#39;click&#39;, function() {<br>            const elementPopup = document.getElementById(&quot;popup-sales-stage&quot;);<br>            elementPopup.open();<br>        });<br>      }<br>    }<br>  }<br><br>  return opportunitiesOnLoadHeaderEventListener;<br>});</pre><h3>Test</h3><ul><li>Run the application and you should see the following:</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*I1RPOcJ551y4HxslIU35OA.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8c98bc2e564f" width="1" height="1" alt=""><hr><p><a href="https://medium.com/oracle-jet/vbs-opportunities-layout-custom-sales-stage-popup-8c98bc2e564f">VBS: Opportunities Layout — Custom Sales Stage Popup</a> was originally published in <a href="https://medium.com/oracle-jet">Oracle VB Studio and JET</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[VB: Business Object API Security]]></title>
            <link>https://medium.com/oracle-jet/vb-business-object-api-security-8bcf099cb650?source=rss-f49c260596a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/8bcf099cb650</guid>
            <category><![CDATA[oracle-cloud]]></category>
            <category><![CDATA[oracle]]></category>
            <category><![CDATA[oracle-visual-builder]]></category>
            <dc:creator><![CDATA[Steve Zebib]]></dc:creator>
            <pubDate>Thu, 23 Jun 2022 04:41:48 GMT</pubDate>
            <atom:updated>2022-06-23T16:58:03.466Z</atom:updated>
            <content:encoded><![CDATA[<p><strong><em>DISCLAIMER: </em></strong><em>The views expressed in this story are my own and do not necessarily reflect the views of Oracle.</em></p><h3>Overview</h3><p>This article describes business object security in Visual Builder. This includes the following:</p><ul><li>Roles: Authenticated User and Anonymous User</li><li>Access Control: Assigning roles to business object API endpoints</li><li>External Access: Accessing business object API outside of Visual Builder using Anonymous User</li></ul><p>Depending on your use case, some business objects may require user authentication rather than anonymous access. Security for each business object in your VB application must be configured separately.</p><h3>Roles</h3><p>The roles associated with business objects include:</p><ul><li>Authenticated User (default): Requires Basic authentication (username/password) to access business object endpoints.</li><li>Anonymous User: Requires Bearer authentication (access token) to access business object endpoints.</li></ul><h3>Access Control</h3><h4><strong>CRUD Operations</strong></h4><p>Furthermore, the following CRUD operations can be assigned one or both of the above roles:</p><ul><li>View: Access to business object GET endpoints.</li><li>Create: Access to business object POST endpoints.</li><li>Update: Access to business object PATCH endpoints.</li><li>Delete: Access to business object DELETE endpoints.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*GHituFqlbkWmA2KQ-kI2gw.png" /><figcaption>Business Object Security Example</figcaption></figure><h4>Environments</h4><ul><li>Development: Only Authenticated User can access business object API endpoints.</li><li>Stage/Live: Authenticated User and/or Anonymous User can access business object API endpoints.</li></ul><h3>External Access</h3><ul><li>To access business object(s) endpoints outside of Visual Builder using Anonymous User following the below configuration.</li></ul><p>NOTE: Business Objects endpoints can be accessed outside of VB only for Staging and Live environments. Development environment CANNOT be accessed outside of VB.</p><h4>Configuration</h4><p><strong>Business Objects</strong></p><ul><li>For this example, we will create a business object called <em>Users </em>as shown below. Go to the Fields tab and add the following fields: name and email.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*GsMw3Fk4xUYjeJsr.png" /></figure><ul><li>Go to the Security tab and check View for Anonymous User as shown below.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*V24sMXPaBtRlC9pA.png" /></figure><ul><li>Go to Endpoints tab and copy the Data Staging URL under Resource APIs. Save URL for later step.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*02B5SivF6F68uV_r.png" /></figure><p><strong>Settings</strong></p><ul><li>Go to Settings (under top-right menu) then click on Business Objects tab. Under Security section, check the following: <strong>Allow anonymous access to business objects describe end point</strong></li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*Hln_tWCTAfL7JuZT.png" /></figure><ul><li>Click on ‘Get Access Token’ button and copy the Access Token Value as shown below. Save access token for later step.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/0*w5UNxSArgBP7kseB.png" /></figure><ul><li>Stage the application.</li></ul><h4>Test</h4><ul><li>Using the Data Staging URL and access token from above steps, we can now make REST calls from outside of VB. We can try this in Postman as shown below.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*40G3ucziZQMnH-KgsvgNmg.png" /></figure><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=8bcf099cb650" width="1" height="1" alt=""><hr><p><a href="https://medium.com/oracle-jet/vb-business-object-api-security-8bcf099cb650">VB: Business Object API Security</a> was originally published in <a href="https://medium.com/oracle-jet">Oracle VB Studio and JET</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[VB: Parsing and Writing Excel Files]]></title>
            <link>https://medium.com/oracle-jet/vb-parsing-and-writing-excel-files-1044d697255d?source=rss-f49c260596a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/1044d697255d</guid>
            <category><![CDATA[visual-builder]]></category>
            <category><![CDATA[saas]]></category>
            <category><![CDATA[oracle-visual-builder]]></category>
            <category><![CDATA[oracle]]></category>
            <category><![CDATA[oracle-cloud]]></category>
            <dc:creator><![CDATA[Steve Zebib]]></dc:creator>
            <pubDate>Wed, 20 Oct 2021 05:49:35 GMT</pubDate>
            <atom:updated>2021-10-20T07:21:41.453Z</atom:updated>
            <content:encoded><![CDATA[<p><strong><em>DISCLAIMER: </em></strong><em>The views expressed in this story are my own and do not necessarily reflect the views of Oracle.</em></p><h3>Overview</h3><p>This article describes how to parse and write Excel files using <a href="https://github.com/sheetjs/sheetjs">SheetJS</a> in Oracle Visual Builder. We will implement the following features in this example:</p><ul><li>JSON to Excel: Parses JSON and generates Excel (.xlsx) file</li><li>Excel to JSON: Generates JSON based on Excel (.xlsx) file</li></ul><p>This tool is specifically useful to import and export data into VB more efficiently by using Excel files.</p><h3>Setup</h3><p><strong>Data</strong></p><ul><li>First, we need sample data to test our implementation. Create a <strong>contacts.json</strong> file under the <strong>resources/data</strong> directory with sample data shown below.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/764/1*lofnqxqTDiD4sqiHyAf_uw.png" /></figure><pre>[<br>  {<br>    &quot;name&quot;: &quot;John Smith&quot;,<br>    &quot;title&quot;: &quot;CEO&quot;,<br>    &quot;company&quot;: &quot;Example&quot;,<br>    &quot;email&quot;: &quot;<a href="/service/mailto:john.smith@example.com">john.smith@example.com</a>&quot;<br>  },<br>  {<br>    &quot;name&quot;: &quot;Lisa Jones&quot;,<br>    &quot;title&quot;: &quot;EVP&quot;,<br>    &quot;company&quot;: &quot;Example&quot;,<br>    &quot;email&quot;: &quot;<a href="/service/mailto:lisa.jones@example.com">lisa.jones@example.com</a>&quot;<br>  },<br>  {<br>    &quot;name&quot;: &quot;Joshua Baker&quot;,<br>    &quot;title&quot;: &quot;CTO&quot;,<br>    &quot;company&quot;: &quot;Example&quot;,<br>    &quot;email&quot;: &quot;<a href="/service/mailto:joshua.baker@example.com">joshua.baker@example.com</a>&quot;<br>  }<br>]</pre><p><strong>SheetJS</strong></p><ul><li>Next, we need to import <a href="https://github.com/SheetJS/sheetjs/blob/master/xlsx.js"><strong>xlsx.js</strong></a><strong> </strong>(from SheetJS)<strong> </strong>into the <strong>resources/js</strong> directory as shown below.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/360/1*L92FHAHbyrJ4DJ87ZVgbsw.png" /></figure><p><strong>JSON</strong></p><ul><li>The Metadata below includes all variables, types, action chains, and events required for this web application. Copy the following into the <strong>main-start-page.json</strong> file:</li></ul><pre>{<br>  &quot;title&quot;: &quot;&quot;,<br>  &quot;description&quot;: &quot;&quot;,<br>  &quot;variables&quot;: {},<br>  &quot;metadata&quot;: {},<br>  &quot;types&quot;: {},<br>  &quot;chains&quot;: {},<br>  &quot;eventListeners&quot;: {<br>    &quot;ojButtonOjAction&quot;: {<br>      &quot;chains&quot;: [<br>        {<br>          &quot;chainId&quot;: &quot;exportExcelActionChain&quot;<br>        }<br>      ]<br>    },<br>    &quot;ojFilePickerOjSelect&quot;: {<br>      &quot;chains&quot;: [<br>        {<br>          &quot;chainId&quot;: &quot;importExcelActionChain&quot;,<br>          &quot;parameters&quot;: {<br>            &quot;files&quot;: &quot;[[ Array.prototype.slice.call($event.detail.files) ]]&quot;<br>          }<br>        }<br>      ]<br>    }<br>  },<br>  &quot;imports&quot;: {<br>    &quot;components&quot;: {<br>      &quot;oj-button&quot;: {<br>        &quot;path&quot;: &quot;ojs/ojbutton&quot;<br>      },<br>      &quot;oj-file-picker&quot;: {<br>        &quot;path&quot;: &quot;ojs/ojfilepicker&quot;<br>      }<br>    }<br>  },<br>  &quot;events&quot;: {}<br>}</pre><p><strong>HTML</strong></p><ul><li>The HTML below includes diagram component with its templates. Copy the following into <strong>main-start-page.html</strong> file:</li></ul><pre>&lt;div class=&quot;oj-flex oj-sm-flex-direction-column oj-sm-align-items-flex-start&quot;&gt;<br> &lt;h4&gt;JSON to Excel&lt;/h4&gt;<br> &lt;oj-button chroming=&quot;callToAction&quot; on-oj-action=&quot;[[$listeners.ojButtonOjAction]]&quot; style=&quot;margin-right: 10px;&quot;&gt;Export<br>    Excel (.xlsx)&lt;/oj-button&gt;<br> &lt;h4&gt;Excel to JSON&lt;/h4&gt;<br> &lt;oj-file-picker select-on=&quot;click&quot; selection-mode=&quot;single&quot;<br>    accept=&#39;[&quot;application/vnd.openxmlformats-officedocument.spreadsheetml.sheet&quot;]&#39;<br>    on-oj-select=&quot;[[$listeners.ojFilePickerOjSelect]]&quot;&gt;<br>  &lt;oj-button id=&quot;oj-button-1225781865-1-import&quot; slot=&#39;trigger&#39; chroming=&quot;callToAction&quot;&gt;Import Excel (.xlsx)<br>    &lt;/oj-button&gt;<br> &lt;/oj-file-picker&gt;<br>&lt;/div&gt;</pre><p><strong>JS</strong></p><ul><li>The JS below includes functions needed to initialize the diagram component. This includes functions which will be assigned to variables that are referenced by the component. Copy the following into <strong>main-start-page.js </strong>file:</li></ul><pre>define([&#39;resources/js/xlsx&#39;, &#39;text!resources/data/contacts.json&#39;], function (XLSX, jsonAsText) {</pre><pre>&#39;use strict&#39;;</pre><pre>// Reference: SheetJS <a href="https://github.com/sheetjs/sheetjs">https://github.com/sheetjs/sheetjs</a></pre><pre>const EXCEL_FILE_NAME = &#39;contacts.xlsx&#39;;<br> const SHEET_NAME = &#39;Sheet1&#39;;</pre><pre>var PageModule = function PageModule() {};</pre><pre>PageModule.prototype.exportExcel = function() {<br>  let dataArray = JSON.parse(jsonAsText);</pre><pre>// Create a new workbook<br>  var wb = XLSX.utils.book_new();</pre><pre>// Assign workbook properties<br>  wb.Props = {<br>   Title: &quot;Example&quot;,<br>   Subject: &quot;&quot;,<br>   Author: &quot;Example&quot;,<br>   CreatedDate: new Date()<br>  };</pre><pre>// Create worksheet<br>  wb.SheetNames.push(SHEET_NAME);</pre><pre>// Convert JSON to worksheet<br>  var ws = XLSX.utils.json_to_sheet(dataArray);</pre><pre>// Assign sheet to workbook<br>  wb.Sheets[SHEET_NAME] = ws;</pre><pre>// Create Excel file<br>  XLSX.writeFile(wb, EXCEL_FILE_NAME);<br> };</pre><pre>PageModule.prototype.importExcel = function(fileSet) {<br>  var readDataPromise = new Promise(function(resolve) {<br>   if (fileSet.length &gt; 0) {<br>    var excelFile = fileSet[0];<br>    if (excelFile.type === &#39;application/vnd.openxmlformats-officedocument.spreadsheetml.sheet&#39;) {<br>     var fileReader = new FileReader();<br>     fileReader.onload = function(fileReadEvent) {<br>      var data = fileReadEvent.target.result;<br>      parseData(data);<br>     };<br>     fileReader.readAsArrayBuffer(excelFile);<br>    }<br>   }<br>  });</pre><pre>return readDataPromise;<br> };</pre><pre>function parseData(arrayBuffer) {<br>  let dataArray = [];</pre><pre>// Get workbook<br>  var wb = XLSX.read(arrayBuffer, {<br>   type: &#39;array&#39;<br>  });</pre><pre>// Get worksheet<br>  // NOTE: This assumes the default sheet name is &quot;Sheet1&quot;<br>  var ws = wb.Sheets[SHEET_NAME];</pre><pre>if (ws) {<br>   // Convert worksheet to JSON<br>   dataArray = XLSX.utils.sheet_to_json(ws);<br>  }</pre><pre>console.log(dataArray);<br>  return dataArray;<br> }</pre><pre>return PageModule;<br>});</pre><h3>Test</h3><ul><li>Run the application and you should see the following:</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*EyrWmKCp9hkxbFkPOTVDpg.png" /></figure><ul><li>Click on ‘Export Excel’ button. This will parse the JSON data and generate <strong>contacts.xlsx </strong>file.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*nKor2-fPYghqAZ8bO7pIUA.png" /></figure><ul><li>Click on ‘Import Excel’ button and select <strong>contacts.xlsx</strong> (downloaded in previous step). This will read the <strong>contacts.xlsx </strong>file and convert to JSON.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*opVuusiEG3RXR0H_5JXFWA.png" /></figure><h3>References</h3><p><a href="https://github.com/sheetjs/sheetjs">SheetJS</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=1044d697255d" width="1" height="1" alt=""><hr><p><a href="https://medium.com/oracle-jet/vb-parsing-and-writing-excel-files-1044d697255d">VB: Parsing and Writing Excel Files</a> was originally published in <a href="https://medium.com/oracle-jet">Oracle VB Studio and JET</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[VB: Dynamic List View with Drag and Drop]]></title>
            <link>https://medium.com/oracle-jet/vb-dynamic-list-view-with-drag-and-drop-16441f3fe25b?source=rss-f49c260596a5------2</link>
            <guid isPermaLink="false">https://medium.com/p/16441f3fe25b</guid>
            <category><![CDATA[oracle]]></category>
            <category><![CDATA[oracle-visual-builder]]></category>
            <category><![CDATA[saas]]></category>
            <category><![CDATA[oracle-cloud]]></category>
            <category><![CDATA[visual-builder]]></category>
            <dc:creator><![CDATA[Steve Zebib]]></dc:creator>
            <pubDate>Fri, 06 Aug 2021 01:00:57 GMT</pubDate>
            <atom:updated>2021-10-20T05:52:21.543Z</atom:updated>
            <content:encoded><![CDATA[<p><strong><em>DISCLAIMER: </em></strong><em>The views expressed in this story are my own and do not necessarily reflect the views of Oracle.</em></p><h3>Overview</h3><p>This article provides an example for how to generate multiple list view components with drag and drop enabled. The drag and drop is useful for moving items between different list views.</p><p>For this example, we will build a drag and drop dashboard for moving tasks from one status to another. This will include 3 separate lists one for each of the following statuses: Open, In Progress, and Closed. This example will also generate the list views dynamically when the application starts.</p><h3>Setup</h3><p>NOTE: All of the below files including the JSON, HTML, and JS can be found in the VBCS web application main page flow.</p><p><strong>JSON</strong></p><ul><li>The Metadata below includes all variables, types, action chains, and events required for this web application. Copy the following into the <strong>main-start-page.json</strong> file:</li></ul><pre>{<br>  &quot;title&quot;: &quot;&quot;,<br>  &quot;description&quot;: &quot;&quot;,<br>  &quot;variables&quot;: {<br>    &quot;dataReady&quot;: {<br>      &quot;type&quot;: &quot;boolean&quot;,<br>      &quot;defaultValue&quot;: false<br>    },<br>    &quot;treeDataProvider&quot;: {<br>      &quot;type&quot;: &quot;any&quot;<br>    }<br>  },<br>  &quot;metadata&quot;: {},<br>  &quot;types&quot;: {},<br>  &quot;chains&quot;: {<br>    &quot;initActionChain&quot;: {<br>      &quot;variables&quot;: {},<br>      &quot;root&quot;: &quot;callFunctionBuildTree&quot;,<br>      &quot;actions&quot;: {<br>        &quot;callFunctionBuildTree&quot;: {<br>          &quot;module&quot;: &quot;vb/action/builtin/callModuleFunctionAction&quot;,<br>          &quot;parameters&quot;: {<br>            &quot;module&quot;: &quot;[[ $functions ]]&quot;,<br>            &quot;functionName&quot;: &quot;buildTree&quot;,<br>            &quot;returnType&quot;: &quot;any&quot;,<br>            &quot;params&quot;: []<br>          },<br>          &quot;outcomes&quot;: {<br>            &quot;success&quot;: &quot;assignVariablesTreeDataProvider&quot;<br>          }<br>        },<br>        &quot;assignVariablesReady&quot;: {<br>          &quot;module&quot;: &quot;vb/action/builtin/assignVariablesAction&quot;,<br>          &quot;parameters&quot;: {<br>            &quot;$page.variables.dataReady&quot;: {<br>              &quot;source&quot;: true<br>            }<br>          }<br>        },<br>        &quot;assignVariablesTreeDataProvider&quot;: {<br>          &quot;module&quot;: &quot;vb/action/builtin/assignVariablesAction&quot;,<br>          &quot;parameters&quot;: {<br>            &quot;$page.variables.treeDataProvider&quot;: {<br>              &quot;source&quot;: &quot;{{ $chain.results.callFunctionBuildTree }}&quot;,<br>              &quot;reset&quot;: &quot;empty&quot;<br>            }<br>          },<br>          &quot;outcomes&quot;: {<br>            &quot;success&quot;: &quot;assignVariablesReady&quot;<br>          }<br>        }<br>      }<br>    }<br>  },<br>  &quot;eventListeners&quot;: {<br>    &quot;vbEnter&quot;: {<br>      &quot;chains&quot;: [<br>        {<br>          &quot;chainId&quot;: &quot;initActionChain&quot;,<br>          &quot;parameters&quot;: {}<br>        }<br>      ]<br>    }<br>  },<br>  &quot;imports&quot;: {<br>    &quot;components&quot;: {<br>      &quot;oj-input-date&quot;: {<br>        &quot;path&quot;: &quot;ojs/ojdatetimepicker&quot;<br>      },<br>      &quot;oj-list-item-layout&quot;: {<br>        &quot;path&quot;: &quot;ojs/ojlistitemlayout&quot;<br>      },<br>      &quot;oj-list-view&quot;: {<br>        &quot;path&quot;: &quot;ojs/ojlistview&quot;<br>      }<br>    }<br>  },<br>  &quot;events&quot;: {}<br>}</pre><p><strong>HTML</strong></p><ul><li>The HTML below includes diagram component with its templates. Copy the following into <strong>main-start-page.html</strong> file:</li></ul><pre>&lt;oj-bind-if test=&quot;[[ $variables.dataReady ]]&quot;&gt;<br> &lt;div id=&quot;container&quot; class=&quot;oj-flex oj-sm-flex-wrap-nowrap&quot;&gt;<br>  &lt;oj-bind-for-each data=&quot;[[ $variables.treeDataProvider ]]&quot;&gt;<br>   &lt;template&gt;<br>    &lt;div class=&quot;oj-flex oj-sm-flex-direction-column&quot; style=&quot;border-right: 1px solid lightgrey;&quot;&gt;<br>     &lt;div<br>      class=&quot;oj-flex oj-sm-align-items-center oj-sm-align-self-center stage-header oj-sm-justify-content-space-between oj-sm-flex-wrap-nowrap&quot;&gt;<br>      &lt;div class=&quot;oj-typography-body-md oj-typography-bold stage-header-text&quot;&gt;<br>       &lt;oj-bind-text<br>        value=&quot;[[ $current.data.statusName + &#39; (&#39; + $functions.getCount($current.data.items) + &#39;) &#39; ]]&quot;&gt;&lt;/oj-bind-text&gt;<br>      &lt;/div&gt;<br>     &lt;/div&gt;<br>     &lt;div class=&quot;oj-flex oj-sm-flex-direction-column oj-sm-flex-wrap-nowrap oj-sm-align-items-center&quot;&gt;<br>      &lt;div class=&quot;oj-flex oj-sm-flex-direction-column&quot;&gt;<br>       &lt;oj-list-view :id=&quot;[[ $current.data.statusId ]]&quot;<br>        data=&quot;[[$functions.buildADP($current.data.statusId,$current.data.items)]]&quot;<br>        dnd.drop.items.data-types=&#39;[&quot;application/ojlistviewitems+json&quot;]&#39;<br>        dnd.drop.items.drop=&quot;[[$functions.handleDrop($current.data.statusId)]]&quot;<br>        dnd.drag.items.data-types=&#39;[&quot;application/ojlistviewitems+json&quot;]&#39;<br>        dnd.drag.items.drag-start=&quot;[[$functions.handleDragStart($current.data.statusId)]]&quot;<br>        dnd.drag.items.drag-end=&quot;[[$functions.handleDragEnd($current.data.statusId)]]&quot;<br>        style=&quot;min-height: 500px; min-width: 250px;&quot;&gt;<br>        &lt;template slot=&quot;itemTemplate&quot; data-oj-as=&quot;item&quot;&gt;<br>         &lt;li&gt;<br>          &lt;oj-list-item-layout&gt;<br>           &lt;div<br>            class=&quot;oj-panel oj-flex oj-sm-flex-direction-column oj-sm-justify-content-space-between opty-card&quot;&gt;<br>            &lt;div<br>             class=&quot;oj-flex oj-sm-align-items-center oj-sm-justify-content-center&quot;&gt;<br>             &lt;div role=&quot;button&quot; tabindex=&quot;0&quot;<br>              class=&quot;oj-sm-align-self-center oj-sm-margin-4x-horizontal oj-listview-drag-handle&quot;&gt;&lt;/div&gt;<br>            &lt;/div&gt;<br>            &lt;div class=&quot;oj-flex&quot; style=&quot;padding: 5px&quot;&gt;<br>             &lt;div class=&quot;oj-text-color-primary&quot;&gt;<br>              &lt;oj-bind-text value=&quot;[[ item.data.Name  ]]&quot;&gt;&lt;/oj-bind-text&gt;<br>             &lt;/div&gt;<br>            &lt;/div&gt;<br>            &lt;div class=&quot;oj-flex-bar oj-sm-align-items-center&quot;<br>             style=&quot;height:40px; padding:5px&quot;&gt;<br>             &lt;div class=&quot;oj-flex oj-sm-flex-direction-column oj-flex-bar-start&quot;&gt;<br>              &lt;div<br>               class=&quot;oj-flex-item oj-sm-12 oj-text-color-secondary oj-md-12&quot;&gt;<br>               &lt;oj-input-date <br>                value=&quot;[[ item.data.EffectiveDate ]]&quot; readonly=&quot;true&quot;<br>                disabled=&quot;true&quot;&gt;&lt;/oj-input-date&gt;<br>              &lt;/div&gt;<br>             &lt;/div&gt;<br>             &lt;div class=&quot;oj-flex oj-sm-flex-direction-column oj-flex-bar-end&quot;&gt;<br>              &lt;span class=&quot;oj-flex-item oj-sm-flex-initial vb-icon vb-icon-ellipsis&quot;&gt;&lt;/span&gt;<br>             &lt;/div&gt;<br>            &lt;/div&gt;<br>           &lt;/div&gt;<br>          &lt;/oj-list-item-layout&gt;<br>         &lt;/li&gt;<br>        &lt;/template&gt;<br>       &lt;/oj-list-view&gt;<br>      &lt;/div&gt;<br>     &lt;/div&gt;<br>    &lt;/div&gt;<br>   &lt;/template&gt;<br>  &lt;/oj-bind-for-each&gt;<br> &lt;/div&gt;<br>&lt;/oj-bind-if&gt;</pre><p><strong>JS</strong></p><ul><li>The JS below includes functions needed to initialize the diagram component. This includes functions which will be assigned to variables that are referenced by the component. Copy the following into <strong>main-start-page.js </strong>file:</li></ul><pre>define([&quot;require&quot;, &quot;knockout&quot;, &quot;ojs/ojbootstrap&quot;, &quot;ojs/ojarraytreedataprovider&quot;, &quot;ojs/ojarraydataprovider&quot;, &quot;ojs/ojknockout&quot;, &quot;ojs/ojlistviewdnd&quot;], function(<br> require, ko, ojbootstrap_1, ArrayTreeDataProvider, ArrayDataProvider) {<br> &#39;use strict&#39;;</pre><pre>var treeDataArray;<br> var statusMap = new Map();<br> var dataTypes = &quot;application/ojlistviewitems+json&quot;;<br> var dataArrayMap = new Map();<br> var handleDragStartMap = new Map();<br> var handleDragEndMap = new Map();<br> var handleDropMap = new Map();</pre><pre>var statusArray = [{<br>   &quot;Id&quot;: 1,<br>   &quot;Name&quot;: &quot;Open&quot;<br>  },<br>  {<br>   &quot;Id&quot;: 2,<br>   &quot;Name&quot;: &quot;In Progress&quot;<br>  },<br>  {<br>   &quot;Id&quot;: 3,<br>   &quot;Name&quot;: &quot;Closed&quot;<br>  }<br> ];</pre><pre>var dataArray = [{<br>   &quot;Name&quot;: &quot;Task #1&quot;,<br>   &quot;Id&quot;: 10001,<br>   &quot;EffectiveDate&quot;: &quot;2021-08-01T00:00:00.000+0000&quot;,<br>   &quot;StatusId&quot;: 3<br>  },<br>  {<br>   &quot;Name&quot;: &quot;Task #2&quot;,<br>   &quot;Id&quot;: 10002,<br>   &quot;EffectiveDate&quot;: &quot;2021-08-01T00:00:00.000+0000&quot;,<br>   &quot;StatusId&quot;: 3<br>  },<br>  {<br>   &quot;Name&quot;: &quot;Task #3&quot;,<br>   &quot;Id&quot;: 10003,<br>   &quot;EffectiveDate&quot;: &quot;2021-08-01T00:00:00.000+0000&quot;,<br>   &quot;StatusId&quot;: 2<br>  },<br>  {<br>   &quot;Name&quot;: &quot;Task #4&quot;,<br>   &quot;Id&quot;: 10004,<br>   &quot;EffectiveDate&quot;: &quot;2021-08-01T00:00:00.000+0000&quot;,<br>   &quot;StatusId&quot;: 2<br>  },<br>  {<br>   &quot;Name&quot;: &quot;Task #5&quot;,<br>   &quot;Id&quot;: 10005,<br>   &quot;EffectiveDate&quot;: &quot;2021-08-01T00:00:00.000+0000&quot;,<br>   &quot;StatusId&quot;: 1<br>  }<br> ];</pre><pre>var PageModule = function PageModule() {};</pre><pre>PageModule.prototype.buildTree = function() {<br>  for (let i = 0; i &lt; statusArray.length; i++) {<br>   let status = statusArray[i];<br>   if (!statusMap.has(status.Id)) {<br>    let data = {<br>     &quot;statusId&quot;: status.Id,<br>     &quot;statusName&quot;: status.Name,<br>     &quot;items&quot;: []<br>    };</pre><pre>statusMap.set(status.Id, data);<br>   }<br>  };</pre><pre>treeDataArray = processArray(dataArray);</pre><pre>return new ArrayTreeDataProvider(treeDataArray, {<br>   keyAttributes: &#39;id&#39;,<br>   keyAttributesScope: &#39;siblings&#39;,<br>   childrenAttribute: &#39;items&#39;<br>  });<br> };</pre><pre>function processArray(array) {<br>  if (array) {<br>   for (let i = 0; i &lt; array.length; i++) {<br>    let item = array[i];</pre><pre>if (item) {<br>     let data = statusMap.get(item.StatusId);<br>     data.items.push(item);<br>     data.count = data.items.length;</pre><pre>statusMap.set(item.StatusId, data);<br>    }<br>   };<br>  };</pre><pre>let updatedArray = [...statusMap.values()];<br>  updatedArray.sort((a, b) =&gt; (a.statusId &gt; b.statusId) ? 1 : -1);</pre><pre>return ko.observableArray(updatedArray);<br> };</pre><pre>PageModule.prototype.getCount = function(array) {<br>  let count = 0;</pre><pre>if (array) {<br>   count = array.length;<br>  }</pre><pre>return count;<br> };</pre><pre>PageModule.prototype.buildADP = function(statusId, array) {<br>  this.dataArray = ko.observableArray(array);<br>  dataArrayMap.set(statusId, this.dataArray);</pre><pre>let dataProvider = new ArrayDataProvider(this.dataArray, {<br>   keyAttributes: &#39;Id&#39;<br>  });</pre><pre>return dataProvider;<br> };</pre><pre>PageModule.prototype.handleDragStart = function(statusId) {<br>  // Handle drag start from drag and drop with mouse/touch.<br>  let handleDragStart = (event, context) =&gt; {<br>   const dataStr = event.dataTransfer.getData(dataTypes);<br>   const data = JSON.parse(dataStr);<br>   this.sourceItemId = context.items[0].Id;<br>   this.sourceStatusId = context.items[0].StatusId;<br>  };</pre><pre>handleDragStartMap.set(statusId, handleDragStart);<br>  return handleDragStart;<br> };</pre><pre>PageModule.prototype.handleDragEnd = function(statusId) {<br>  // Handle drag end from drag and drop with mouse/touch.<br>  let handleDragEnd = (event) =&gt; {<br>   if (event.dataTransfer.dropEffect !== &quot;none&quot;) {}<br>  };</pre><pre>handleDragEndMap.set(statusId, handleDragEnd);<br>  return handleDragEnd;<br> };</pre><pre>PageModule.prototype.handleDrop = function(statusId) {<br>  let handleDrop = (event, context) =&gt; {<br>   event.preventDefault();<br>   let index = -1;<br>   if (context.item) {<br>    const itemContext = document.getElementById(statusId).getContextByNode(context.item);<br>    index = itemContext.index;<br>    if (context.position === &quot;after&quot;) {<br>     index += 1;<br>    }<br>   }</pre><pre>handleDataTransfer(event.dataTransfer, index, statusId);<br>  };</pre><pre>// Handle drop from drag and drop,<br>  let handleDataTransfer = (dataTransfer, index, statusId) =&gt; {<br>   const dataStr = dataTransfer.getData(dataTypes);<br>   const data = JSON.parse(dataStr)[0];<br>   if (this.sourceStatusId != statusId) {<br>    insertTargetItem(data, index, statusId);<br>    removeSourceItem(this.sourceItemId, this.sourceStatusId);<br>   }<br>  };</pre><pre>// Insert item into array.  This is invoked by drop.<br>  let insertTargetItem = (data, index, targetStatusId) =&gt; {<br>   const arr = dataArrayMap.get(targetStatusId);</pre><pre>data.StatusId = targetStatusId;</pre><pre>if (index === -1) {<br>    arr.push(data);<br>   } else {<br>    arr.splice(index, 0, data);<br>   }</pre><pre>arr.valueHasMutated();<br>   treeDataArray.valueHasMutated();<br>  };</pre><pre>// Remove item from array.  This is invoked by drop.<br>  let removeSourceItem = (itemId, sourceStatusId) =&gt; {<br>   const arr = dataArrayMap.get(sourceStatusId);<br>   if (arr) {<br>    arr.remove(function(item) {<br>     return item.Id === itemId;<br>    });<br>    arr.valueHasMutated();<br>    treeDataArray.valueHasMutated();<br>   }<br>  };</pre><pre>handleDropMap.set(statusId, handleDrop);<br>  return handleDrop;<br> };</pre><pre>return PageModule;<br>});</pre><h3>Test</h3><ul><li>Run the application and you should see the following:</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*vEYNx89cGtWDhrvFR-7W7A.png" /></figure><ul><li>Drag and drop an item from one status to another. For example, we will move Task #3 from ‘In Progress’ to ‘Closed’ status as shown below.</li></ul><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*xBCOyagJ2pVyXXaN663iag.png" /></figure><h3>References</h3><p><a href="https://www.oracle.com/webfolder/technetwork/jet/jetCookbook.html?component=listView&amp;demo=withTableDndListView">Oracle JET Cookbook — List View Drag and Drop</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=16441f3fe25b" width="1" height="1" alt=""><hr><p><a href="https://medium.com/oracle-jet/vb-dynamic-list-view-with-drag-and-drop-16441f3fe25b">VB: Dynamic List View with Drag and Drop</a> was originally published in <a href="https://medium.com/oracle-jet">Oracle VB Studio and JET</a> on Medium, where people are continuing the conversation by highlighting and responding to this story.</p>]]></content:encoded>
        </item>
    </channel>
</rss>