DEV Community: Naveen

AI-Powered BI Publisher Report Retrieval in Oracle Fusion Agent AI 26C

Naveen — Wed, 12 Aug 2026 08:37:41 +0000

Over the last few days, I designed and experimented with an exciting use case using Oracle Fusion Agent AI Studio 26C, combining External REST Tools, Agent Calls and Code Nodes to bring enterprise reporting directly into a conversational AI experience.

Project Vision

The goal was to enable users to ask business questions in natural language and have Oracle Fusion Agent AI:

Invoke a BI Publisher Report on demand
Retrieve report output via REST APIs
Process the response dynamically
Convert report data into a readable format
Present business insights directly within the chat conversation

Example Queries

"Show me the latest supplier invoices."
or
"Display open purchase orders for my business unit."

Solution Architecture

To achieve conversational access to Oracle Fusion reports
I designed an architecture that combines Oracle Fusion Agent AI, External REST Tools, BI Publisher Reports, and Agent Calls to deliver report information through a natural language interface.

Solution Architecture: Oracle Fusion Agent AI Studio interface overview

The Below is Tested & Designed Components in Studio

Figure 1: Real Oracle Fusion Agent AI Studio interface overview Figure 2: Agent which Is capable of the Decode ReportBytes & return results

Needed Prompt🤔

The below prompts need to be added in the agent.

Agent Role:

RESPONSE BEHAVIOR

Tone and voice
- Be professional, concise, and neutral — this agent surfaces HR/personal data, not casual conversation.
- Do not editorialize about the person's data (no commentary on tenure, status, personal details, etc.) — present facts only.

Structure of responses
- When data is successfully retrieved: lead with the markdown table, then a one-line summary (row count, person number). No preamble like "Sure, here's the data you asked for."
- When no data is found or the call fails: state plainly what happened and what the user can do next (e.g., re-check the person number, try again shortly). Never show an empty table — say there's no data instead.
- When the person number is missing or ambiguous: ask exactly one direct clarifying question before calling the tool. Do not guess a person number or proceed on an assumption.

Error handling tone
- Never expose internal technical details in a response: no node names, decode statuses, stack traces, SOAP fault XML, or connection identifiers.
- Translate technical failures into plain, actionable language (e.g., "I wasn't able to retrieve that report right now — please try again, or confirm the person number is correct.").
- If the same failure persists after a retry, say so plainly rather than repeating the same generic error silently.

Data handling and privacy
- Only surface the fields relevant to what the user actually asked — do not dump the full record (including sensitive fields like date of birth, marital status, nationality, religion, home address) unless the user specifically asks for that information.
- Never restate a person's sensitive personal fields back for confirmation purposes (e.g., don't echo an address or DOB just to "confirm" a lookup) — confirm using the person number only.
- Do not retain, summarize, or reference a previously retrieved person's data in later turns unless the user re-asks about that same person in the current conversation.

Scope discipline
- Only answer questions this tool can actually address (person lookups by person number). For unrelated requests, state plainly that this assistant is scoped to person report lookups.
- Do not speculate about data the tool didn't return, and do not fill in gaps with assumptions.

Prompt:

You are a helpful assistant with access to the Invoke_BIP_report_NJ tool, which retrieves person details from the XX_PEOPLE_INFO Oracle BI Publisher report.

When a user asks about a person's details or requests information for a specific person number, follow these steps:
1. Extract the Person Number from the user input and call the Invoke_BIP_report_NJ tool with the user-provided value as the Person_Number input parameter.
2. The tool returns reportBytes as base64-encoded content — decode it first, then parse the decoded text as CSV.
3. Use the first row of the decoded CSV as the column headers, and every subsequent row as a data record.
4. Present the data in a clean, well-structured markdown table with appropriate column headers.
5. If the tool call fails, returns no reportBytes, or the person number is not found, inform the user clearly rather than showing an empty or broken table.

Always confirm the person number with the user before calling the tool if it is ambiguous or missing.

When displaying results:
- Format all data as a readable markdown table.
- Use the first row of the decoded CSV as the column headers.
- Align columns cleanly.
- If there are many rows, summarize the total count at the end (e.g., "Showing 1 record for Person Number 1").

This report contains personal information (contact details, date of birth, marital status, nationality, religion). Only display the fields relevant to the user's question rather than the entire record by default, and do not share this data with anyone other than the requesting user.

Create External Tool

Request Body

<soap:Envelope xmlns:soap="http://www.w3.org/2003/05/soap-envelope" xmlns:pub="http://xmlns.oracle.com/oxp/service/PublicReportService">
   <soap:Header/>
   <soap:Body>
      <pub:runReport>
         <pub:reportRequest>
            <pub:attributeFormat>csv</pub:attributeFormat>
            <pub:parameterNameValues>
               <pub:item>
                  <pub:name>P_PERSON_NUM</pub:name>
                  <pub:values>
                     <pub:item>{Person_Number}</pub:item>
                  </pub:values>
               </pub:item>
            </pub:parameterNameValues>
<pub:reportAbsolutePath>/Custom/o_NJ/XX_PEOPLE_INFO.xdo</pub:reportAbsolutePath>          
  <pub:sizeOfDataChunkDownload>-1</pub:sizeOfDataChunkDownload>
         </pub:reportRequest>
      </pub:runReport>
   </soap:Body>
</soap:Envelope>

Challenge Encountered

While I successfully extracted the Base64 encoded report data, decoding the content and rendering the final business data directly within the Agent conversation proved more challenging than expected.

The platform behavior around:

Binary report handling
Base64 decoding
Runtime transformation
Conversational rendering

required deeper exploration than initially anticipated.

One integration, N partners — routed at runtime by dynamic connections

Naveen — Sun, 02 Aug 2026 06:39:58 +0000

Let's DO it Dynamic Connections in OIC & Build Multi-tenant partner router Project

⬇️ click on it

The core idea

Ten partners used to mean ten integrations same logic, copy-pasted, diverging quietly over time.
This build collapses them into one: a single orchestration that decides, per request, which connection to call and which path to call it on, by reading one lookup table.

This works today for FTP and REST adapters

How the resolution mechanism actually works

Every adapter invoke activity that supports this has an Enable Dynamic Connection checkbox in its configuration wizard. Once checked, a new field appears: Dynamic Connection ID.

At runtime, OIC evaluates whatever expression you put there and uses that value to pick which saved connection to actually use for this specific invocation overriding the design-time default.

Two ways to populate that field:

Direct XPath drag a field straight from the incoming payload (e.g., a partnerId element) if the connection ID is literally present in your data.
*Lookup table (DVM) * resolve a business key (partner ID, tenant ID, environment name) to a connection ID via a lookup table, using the lookupValue() function:

lookupValue("PartnerRoutingTable", "PartnerId", partnerId, "ConnectionId", "")

The project: a multi-tenant partner router

To make this concrete, I built and tested an integration that takes an inbound order and routes it to one of three partners — two over FTP, one over REST — using a single lookup table to decide both the connection and the destination path.

Connections:

The routing table(Lookup)

PartnerId	ConnectionId	TargetPath	AdapterType
PARTNER_A	FTP_PartnerA_Conn	/	FTP
PARTNER_B	REST_PartnerB_Conn	/orders	REST
PARTNER_C	FTP_PartnerC_Conn	/	FTP

This table is the only partner-specific knowledge in the whole system. Everything else is generic.

The flow

Trigger (REST, receives order + partnerId)
        │
        ▼
Assign: lookupValue(partnerId) → adapterType, connectionId, targetPath
        │
        ▼
Switch on adapterType
   ├── "FTP"  → Map → Invoke (dynamic connection) → Map
   ├── "REST" → Map → Invoke (dynamic connection) → Map
   └── otherwise → fault: unknown partner
        │
        ▼
   Return response

TEST Results:

Let Pass "partnerId": "PARTNER_C" in request payload

As per partnerId it invoked the C FTP.

Let Pass "partnerId": "PARTNER_B" in request payload

As expected, it went to REST with same dynamic connection.

Requets Payload:

{
  "totalAmount": 99.95,
  "orderId": "ORD-10023",
  "partnerId": "PARTNER_B",
  "orderDate": "2026-08-02",
  "items": [{
    "unitPrice": 19.99,
    "quantity": 5,
    "description": "Sample Widget",
    "sku": "SKU-1001"
  }],
  "customerName": "Test Customer"
}

Lessons from actually building and testing this

The reference docs don't spell out, discovered by hitting them:

ConnectionId in your lookup must match the connection's name exactly. It's resolved as a literal string at runtime, not validated against existing connections at design time — a typo here fails silently in production, not on save.

Where else this pattern applies

The router doesn't actually know it's routing "partners" — it only knows "resolve a lookup row, then pick a connection and a path." Swap what the row contains and the same skeleton solves:

Credential rotation — one lookup key, two connections (old/new creds); flip the row to cut over instantly
Environment promotion — key the lookup by environment instead of partner; promote dev → test → prod by editing a row, not maintaining three integrations
DR/failover — a health-check flag in the lookup redirects traffic to a standby endpoint the moment it's toggled
Tiered routing — premium accounts resolve to a dedicated connection, everyone else shares one

Building a Real-Time Collaborative Notes App on Oracle Backend with Firebase APIs

Naveen — Fri, 31 Jul 2026 09:41:48 +0000

Author: Naveen Jejji
Published: July 31, 2026

The project

Collaborative Notes is a small web app where signed-in users share a single live notebook: anyone can create a note, anyone signed in can edit anyone else's note, and edits appear for every other open tab or user instantly no refresh button, no polling. Notes can carry an image or PDF attachment with an inline preview. Only the original creator of a note can delete it.

It's a deliberately "boring" app on the surface notes, not rocket science chosen specifically because real-time multi-user collaboration, file uploads and ownership-based permissions together are a genuine stress test for a backend, not a toy demo.

Repo (full source)

The Oracle technology behind it

The backend is Oracle Backend with Firebase APIs a free feature of Oracle REST Data Services (ORDS) that lets you write against Oracle AI Database using the same open-source Firebase client SDKs developers already know: authentication, a Firestore-shaped document API, object storage, and declarative security rules.

There's no separate backend service to stand up and no additional license — if you're already running ORDS, this rides on top of it.

That matters for this specific app in three concrete ways.

1. Real-time sync via `onSnapshot`

The entire "collaborative" feel of the app comes from a single live query listener instead of a fetch-on-demand call:

function startNotesListener() {
  const notesQuery = query(collection(db, "notes"), orderBy("updatedAt", "desc"), limit(100));

  onSnapshot(
    notesQuery,
    (snapshot) => {
      state.notes = snapshot.docs.map((d) => ({ id: d.id, ...d.data() }));
      renderNotes();
    },
    (error) => {
      renderStatus(error.message, "error");
    }
  );
}

Every create, edit, and delete from any user, any tab refires this callback automatically.
The UI just re-renders from whatever the current snapshot is. There's no manual "refresh" logic anywhere in the app for the notes list.

2. File uploads with inline preview

Attaching an image or PDF to a note uses the same Storage API shape as Firebase Storage upload, then resolve a download URL:

async function uploadAttachment(noteId, file) {
  const fileRef = ref(storage, `notes/${noteId}/${file.name}`);
  await uploadBytes(fileRef, file, { contentType: file.type || "application/octet-stream" });
  const url = await getDownloadURL(fileRef);
  return { attachmentURL: url, attachmentType: file.type, attachmentName: file.name };
}

The app then renders an <img> thumbnail for images or an inline <embed> preview for PDFs, so attachments are visible in the shared list without leaving the page.

3. A collaborative permission model, enforced server-side

This is the detail I think is most worth highlighting. "Collaborative" here specifically means: any signed-in user can edit any note, but only the creator can delete it. That's not a client-side convention I'm trusting the UI to respect — it's a security rule evaluated on the backend, independent of the JavaScript running in someone's browser:

Create: allowed if the signed-in user is the one being set as the note's owner
Update: allowed for any signed-in user (this is what makes it collaborative rather than single-owner)
Delete: allowed only if the requesting user's ID matches the note's owner field

A user can't just edit the DOM or replay a request to bypass this the rule lives next to the data, not in the page.

What users actually get out of this

No merge conflicts to think about. Because everyone sees the current state live, the odds of two people editing stale versions of the same note drop sharply compared to a "refresh to see changes" model.
Real collaboration, not just shared storage. Anyone can pick up and extend anyone else's note closer to a shared whiteboard than a personal to-do list.
Attachments that are actually usable, not just links you have to download and open separately.
A permission model that's actually enforced, not just implied by what buttons happen to be visible in the UI.
Zero additional backend cost or infrastructure for teams already running Oracle Database + ORDS this is additive, not a new system to operate.

How it's deployed

The backend runs as two containers (Oracle Database Free 26ai + ORDS) via Podman, with the frontend as a plain static site (HTML/CSS/vanilla JS with ES module import maps no build step, no framework).

The static UI is also published via GitHub Pages for browsing the interface live data operations require the Oracle Backend with Firebase APIs instance to be reachable, same as any app that talks to its own backend.

Try it / read the code

Source: https://github.com/naveen-6735/collaborative-notes-web
The workshop this environment is built on: RecipeShare — Oracle LiveLabs (Web)
Oracle Backend with Firebase APIs — landing page
Announcement post
Developer docs

If you build something with this toolkit, I'd genuinely like to hear about it.

Using PL/SQL in Oracle Fusion BI Publisher Reports for Dynamic Data Filtering

Naveen — Sun, 19 Jul 2026 07:14:19 +0000

Oracle Fusion BI Publisher supports PL/SQL Data Sets, allowing report developers to implement business logic directly within a report.

This approach is useful when report data needs to be dynamically filtered or transformed based on user input.

Requirement

In this example, users can select a Supplier Type parameter from the following values:

SUPPLIER
PUBLIC SECTOR COMPANIES
INSURANCE
INTERCOMPANY

Based on the selected value, the report retrieves matching supplier records from the Fusion Procurement supplier table (POZ_SUPPLIERS_V)

Click to view the full PL/SQL Data Set code

DECLARE

    ----------------------------------------------------------------------------
    -- Define a REF CURSOR type.
    -- BI Publisher uses a cursor to return the final result set back to
    -- the report output.
    ----------------------------------------------------------------------------
    TYPE refcursor IS REF CURSOR;

    ----------------------------------------------------------------------------
    -- Cursor variable that will hold the query result.
    -- BI Publisher expects the data to be returned through xdo_cursor.
    ----------------------------------------------------------------------------
    xdo_cursor REFCURSOR;

    ----------------------------------------------------------------------------
    -- Variable used to store the supplier type returned by the function.
    ----------------------------------------------------------------------------
    l_supplier_filter VARCHAR2(100);

    ----------------------------------------------------------------------------
    -- Local PL/SQL Function: Supplier_type
    --
    -- Purpose:
    --   Accepts the report parameter P_VENDOR_TYPE and validates the value.
    --
    -- Allowed Values:
    --   SUPPLIER
    --   PUBLIC SECTOR COMPANIES
    --   INSURANCE
    --   INTERCOMPANY
    --
    -- Returns:
    --   The corresponding supplier type value.
    --   If no valid value is found, returns 'NA'.
    ----------------------------------------------------------------------------
    FUNCTION Supplier_type (
        p_in_vendor_type IN VARCHAR2
    )
    RETURN VARCHAR2
    IS
        ----------------------------------------------------------------------------
        -- Variable to hold the function result.
        ----------------------------------------------------------------------------
        l_supplier_type_rtn VARCHAR2(100);

    BEGIN

        ----------------------------------------------------------------------------
        -- Check if the selected supplier type is SUPPLIER.
        ----------------------------------------------------------------------------
        IF p_in_vendor_type = 'SUPPLIER' THEN

            l_supplier_type_rtn := 'SUPPLIER';

        ----------------------------------------------------------------------------
        -- Check if the selected supplier type is PUBLIC SECTOR COMPANIES.
        ----------------------------------------------------------------------------
        ELSIF p_in_vendor_type = 'PUBLIC SECTOR COMPANIES' THEN

            l_supplier_type_rtn := 'PUBLIC SECTOR COMPANIES';

        ----------------------------------------------------------------------------
        -- Check if the selected supplier type is INSURANCE.
        ----------------------------------------------------------------------------
        ELSIF p_in_vendor_type = 'INSURANCE' THEN

            l_supplier_type_rtn := 'INSURANCE';

        ----------------------------------------------------------------------------
        -- Check if the selected supplier type is INTERCOMPANY.
        ----------------------------------------------------------------------------
        ELSIF p_in_vendor_type = 'INTERCOMPANY' THEN

            l_supplier_type_rtn := 'INTERCOMPANY';

        ----------------------------------------------------------------------------
        -- If none of the above values match, return NA.
        ----------------------------------------------------------------------------
        ELSE

            l_supplier_type_rtn := 'NA';

        END IF;

        ----------------------------------------------------------------------------
        -- Return the validated supplier type value.
        ----------------------------------------------------------------------------
        RETURN l_supplier_type_rtn;

    EXCEPTION

        ----------------------------------------------------------------------------
        -- Handle any unexpected errors and return NA.
        ----------------------------------------------------------------------------
        WHEN OTHERS THEN

            RETURN 'NA';

    END Supplier_type;

BEGIN

    ----------------------------------------------------------------------------
    -- Pass the BI Publisher report parameter P_VENDOR_TYPE
    -- into the Supplier_type function.
    --
    -- Example Input:
    --   SUPPLIER
    --   INSURANCE
    --   PUBLIC SECTOR COMPANIES
    --   INTERCOMPANY
    --
    -- Example Output:
    --   SUPPLIER
    --   INSURANCE
    --   PUBLIC SECTOR COMPANIES
    --   INTERCOMPANY
    ----------------------------------------------------------------------------
    l_supplier_filter := Supplier_type(:P_VENDOR_TYPE);

    ----------------------------------------------------------------------------
    -- Open the cursor and return the result set to BI Publisher.
    ----------------------------------------------------------------------------
    OPEN :xdo_cursor FOR

        ----------------------------------------------------------------------------
        -- Retrieve supplier information from Oracle Fusion Procurement
        -- Supplier View (POZ_SUPPLIERS_V).
        ----------------------------------------------------------------------------
        SELECT
               ----------------------------------------------------------------------------
               -- Supplier Name
               ----------------------------------------------------------------------------
               vendor_name,

               ----------------------------------------------------------------------------
               -- Supplier Type
               ----------------------------------------------------------------------------
               vendor_type_lookup_code,

               ----------------------------------------------------------------------------
               -- Organization Type
               ----------------------------------------------------------------------------
               organization_type_lookup_code,

               ----------------------------------------------------------------------------
               -- Business Relationship
               ----------------------------------------------------------------------------
               business_relationship,

               ----------------------------------------------------------------------------
               -- Display the selected parameter value in the report output.
               ----------------------------------------------------------------------------
               l_supplier_filter supplier_type

        FROM   poz_suppliers_v

        ----------------------------------------------------------------------------
        -- Filter suppliers based on the selected Vendor Type parameter.
        --
        -- Example:
        -- If P_VENDOR_TYPE = 'INSURANCE'
        -- Then the query becomes:
        --
        -- WHERE vendor_type_lookup_code = 'INSURANCE'
        ----------------------------------------------------------------------------
        WHERE  vendor_type_lookup_code = l_supplier_filter;

END;
/

Create Data model in Fusion

As below

Preview the data by passing parameters value.

Save sample data & move to create a report

Conclusion

This PL/SQL-based BI Publisher Data Set demonstrates how Oracle Fusion report developers can combine report parameters, custom functions, and cursor-based queries to create dynamic reports.

By processing the user-selected supplier type through a PL/SQL function and applying it to the query filter, the report becomes more flexible, maintainable, and aligned with business requirements.

This technique is especially useful in Procurement reporting scenarios where supplier classifications drive reporting outcomes.

Real-Time Reporting: How to Push ATP Data to Oracle Fusion BI Publisher

Naveen — Sun, 19 Jul 2026 02:51:45 +0000

Most Oracle Fusion implementations follow a standard pattern: transferring data from Fusion Applications into Oracle Autonomous Transaction Processing (ATP) for analytics

But what happens when you need to go the other way ?

In this project, I designed a solution for a common business pain point: fetching processed data from ATP, generating a dynamic BI Publisher report, and distributing it via Oracle Integration Cloud (OIC) notifications.

The Business Challenge

Organizations store massive volumes of processed data in ATP—everything from billing records to custom integration staging tables. While ATP is great for storage, business users need:

Professional, formatted report templates.
Automated, scheduled delivery.
Rich HTML email notifications.

Our solution automates this entire flow, removing manual effort and custom scripting.

Solution Architecture

The flow moves from data storage to user delivery in a seamless loop:

Technologies Used
Oracle Technology Purpose Oracle ATP Source Data Repository OIC Orchestration Layer Fusion BI Publisher Report Design & Execution SOAP Services Remote Report Execution OCI Hosting Platform

Implementation Steps

Step 1: Create Scheduled Integration in OIC

We start with a Scheduled Integration. This removes human dependency and allows for daily, weekly, or custom business intervals.

Step 2: Connect to ATP and Execute SQL

Using the ATP Adapter, we establish a secure connection to fetch the "Active" records.

SELECT 
    * from emp

Step 2.1: ATP Data to Stage File in OIC

Here we used stage file action in OIC to write the ATP data into a file, which further we mapped to SOAP services as content.

Click to see Loaded sample xml file in the stage file

<?xml version="1.0" encoding="UTF-8" ?><root>
   <row>
      <EMPNO>7839</EMPNO>
      <ENAME>KING</ENAME>
      <JOB>PRESIDENT</JOB>
      <MGR/>
      <HIREDATE>1981-11-17T00:00:00.000+00:00</HIREDATE>
      <SAL>5000</SAL>
      <COMM/>
      <DEPTNO>10</DEPTNO>
   </row>
   <row>
      <EMPNO>7698</EMPNO>
      <ENAME>BLAKE</ENAME>
      <JOB>MANAGER</JOB>
      <MGR>7839</MGR>
      <HIREDATE>1981-05-01T00:00:00.000+00:00</HIREDATE>
      <SAL>2850</SAL>
      <COMM/>
      <DEPTNO>30</DEPTNO>
   </row>
</root>

Here the mapping reference of stage writes from ATP

Step 3: Build Fusion BI Publisher Data Model

The incoming ATP data is structured into an XML format. This serves as the data source for our report.

Click to see Magic in the Fusion

Create Data model with the XML FILE

Loaded the same sample XML File which we used earlier

Preview Data & save as Sample Data

Don't forgot save it in folders.

Create report based on above Data model

Keep the report path with you

Step 4: Execute Fusion Run Report SOAP Service

This is the core of the integration. OIC invokes the runReport()
operation via the BI Publisher SOAP service.

Below is the mapper reference!!

Key Insight: The SOAP service is highly flexible. It accepts the XML data, applies your RTF/Excel template, and generates the output in your choice of format (PDF, Excel, CSV).

Click to see How reportBytes converts to DecodeBase64

Using assignment, I converted to DecodeBase64

Step 5: Send OIC HTML Notification

Finally, we use the OIC Notification Action to deliver the report.

Click to see Sample Email Template:

Sample Email Template:

<html>
<body>
<p>Hi Nj</p>
<p>Generate excel file using BIP report<br>
DB data - XML Data Model fusion - report xlsx</p>

<p>{bfy}</p>

<p>Best Regards,<br>Naveen Jejji<br>OIC Dev</p>


<div class="logo">
<img src ="https://osidigital.com/wp-content/uploads/2018/08/osi_logo_rgb.png" alt="company logo"
width ="250">
</div>
</body>
</html>

Key Innovation: Flipping the Script

The Paradigm Shift

Traditional: Fusion → ATP (Data Archiving)
Our Approach: ATP → Fusion BI Publisher → OIC (Dynamic Reporting)

This leverages ATP as the "Source of Truth" while utilizing Fusion’s powerful formatting engine.

Measurable Outcomes

Metric Improvement Manual Effort Reduced by 90% Delivery Time Hours to Minutes Automation 100% Automated

Conclusion

This project demonstrates that the Oracle ecosystem is most powerful when its components work in a "reverse" architecture.

By moving data from ATP back into the Fusion reporting engine, we provide users with the professional outputs they need without manual intervention.

I’d love to hear your thoughts:

Have you implemented a similar "reverse" integration? What challenges did you face with the SOAP service limits? Let's discuss in the comments!

Automating Customer Termination Lifecycle Using Oracle Integration Cloud, ATP, and Oracle Service Cloud

Naveen — Sat, 04 Jul 2026 16:29:34 +0000

Introduction

I built an automated customer termination framework using Oracle Fusion Cloud, OIC, ATP, and Oracle Service Cloud.

It synchronizes termination data across systems, eliminating manual delays, billing errors and operational overhead while ensuring consistent customer lifecycle management.

Solution Architecture

The automated lifecycle flow

Integration Flow

Step 1
Customer termination information is identified in Oracle Fusion.

 Fusion → BIP report with “last run” parameter → OIC → ATP → Service Cloud.

Fusion → BIP report with “last run” parameter → OIC

A Quick Business Validation
The integration invokes Oracle Service Cloud APIs.
I added to check whether customer is available in RightNow or Not using RESTAPI

SerachCustomers in RightNow

Step 2
Based on above response, we used switch in OIC -To handle creation & update of customer in RightNow(Oracle Service Cloud (RightNow) invoke):

Oracle Service Cloud (RightNow) invoke

Business Value Delivered

The solution successfully delivered:

Automated customer termination processing
Improved synchronization between Oracle Fusion and Oracle Service Cloud
Reduced billing risk
Strong audit capabilities
Better support visibility
Enterprise-grade scalability

Conclusion
By leveraging Oracle Integration Cloud, ATP, and Oracle Service Cloud, we successfully automated the customer termination lifecycle and eliminated several manual operational challenges.

The solution demonstrates how Oracle's cloud platform can be used to build reliable, scalable, and business-critical integrations that improve efficiency while maintaining data integrity across enterprise applications.

Stop Using Q! Use Finder for Faster Oracle OIC REST Integrations

Naveen — Sun, 14 Jun 2026 16:36:19 +0000

Learn how to use the Finder parameter in Oracle Fusion REST APIs instead of Q for better performance, cleaner OIC mappings, and optimized integrations.

While Finder works instantly in Postman, OIC requires it to be constructed as a single parameter using XPath expressions, that’s where most developers struggle

In OIC, integrating with Oracle Fusion Various REST APIs is a common requirement for extracting and processing business data. One of the key decisions developers face is how to filter data efficiently using query parameters.

Most developers are familiar with the q parameter, but Oracle also provides a powerful alternative the finder option.

What is finder in Oracle Fusion REST APIs?
The finder parameter is a predefined query mechanism provided by Oracle Fusion REST APIs to retrieve data using named queries and bind parameters

Eg:

...?finder=findPurchaseOrderSchedules;
bindPONumber=AE01-P001531;
bindPOItemNumber=664-8455

It behaves like parameterized SQL queries (bind variables), improving performance and readability.

What is q Parameter?
The q parameter is used to apply dynamic filters similar to a SQL WHERE clause
Eg:

...?q=PONumber='AE01-P001531' and POItemNumber='664-8455'

Flexible but less structured
Can lead to complex expressions

let's Walk into OIC......!

Make sure you have required REST connections.
Create Application based Integration with POST method.

Note:The below fields are we need to pass for another REST API.

Call Get Method with finder using rest call. Below reference, which we will call

Add Finder as review parameter for the endpoint

Now we will map to finder respectively as below.

End flow as below

TESTING

STRUGGLING WITH IRREGULAR CSV FILES IN OIC? HERE’S THE NXSD SOLUTION

Naveen — Wed, 10 Jun 2026 17:14:01 +0000

Handling non-uniform CSV/TXT files in Oracle Integration Cloud (OIC) is a common real-time integration challenge. These files often contain inconsistent structures like headers, mixed record types, or variable columns. This blog explains how to process such files using NXSD (Native Format Schema Definition) and trigger notifications effectively.

UNDERSTANDING THE CHALLENGE

Below is csv/txt file, which is Variable number of columns

Code,Message,Date
F,Hello,nj@gmail.com,957351,01-06-2026
S,Hey,ab@gmail.com,45466,02-06-2026
P,Bye,ck@gmail.com,87979,03-06-2026

This kind of files we need to process in OIC with most effectively.

STEP 1: CREATE FLEXIBLE NXSD SCHEMA
NXSD enables parsing flat files into structured XML.

Note: Observe & understand the below script with respective the challenge mentioned.

<?xml version="1.0" encoding="UTF-8" ?>
<xsd:schema xmlns:xsd="http://www.w3.org/2001/XMLSchema"
            xmlns:nxsd="http://xmlns.oracle.com/pcbpel/nxsd"
            xmlns:tns="http://TargetNamespace.com/fileReference/read"
            targetNamespace="http://TargetNamespace.com/fileReference/read"
            elementFormDefault="qualified"
            attributeFormDefault="unqualified"
            nxsd:version="NXSD"
            nxsd:stream="chars"
            nxsd:encoding="UTF8">

    <xsd:element name="Root-Element">
        <xsd:complexType>
            <xsd:sequence>

                <xsd:element name="headerset">
                    <xsd:complexType>
                        <xsd:sequence>
                            <xsd:element ref="tns:header"/>
                        </xsd:sequence>
                    </xsd:complexType>
                </xsd:element>

                <xsd:element name="lineset">
                    <xsd:complexType>
                        <xsd:sequence>
                            <xsd:element ref="tns:line" minOccurs="1" maxOccurs="unbounded"/>
                        </xsd:sequence>
                    </xsd:complexType>
                </xsd:element>

            </xsd:sequence>
        </xsd:complexType>
    </xsd:element>
    <xsd:element name="header" type="tns:headerType"/>
    <xsd:element name="line" type="tns:lineType"/>
    <xsd:complexType name="headerType">
        <xsd:sequence>
            <xsd:element name="C1" type="xsd:string"
                         nxsd:style="terminated"
                         nxsd:terminatedBy=","
                         nxsd:quotedBy="&quot;"/>
            <xsd:element name="C2" type="xsd:string"
                         nxsd:style="terminated"
                         nxsd:terminatedBy=","
                         nxsd:quotedBy="&quot;"/>
            <xsd:element name="C3" type="xsd:string"
                         nxsd:style="terminated"
                         nxsd:terminatedBy="${eol}"
                         nxsd:quotedBy="&quot;"/>
        </xsd:sequence>
    </xsd:complexType>
    <xsd:complexType name="lineType">
     <xsd:sequence>
            <xsd:element name="C1" type="xsd:string"
                         nxsd:style="terminated"
                         nxsd:terminatedBy=","
                         nxsd:quotedBy="&quot;"/>
            <xsd:element name="C2" type="xsd:string"
                         nxsd:style="terminated"
                         nxsd:terminatedBy=","
                         nxsd:quotedBy="&quot;"/>
            <xsd:element name="C3" type="xsd:string"
                         nxsd:style="terminated"
                         nxsd:terminatedBy=","
                         nxsd:quotedBy="&quot;"/>
            <xsd:element name="C4" type="xsd:string"
                         nxsd:style="terminated"
                         nxsd:terminatedBy=","
                         nxsd:quotedBy="&quot;"/>
            <xsd:element name="C5" type="xsd:string"
                         nxsd:style="terminated"
                         nxsd:terminatedBy="${eol}"
                         nxsd:quotedBy="&quot;"/>               
        </xsd:sequence>
    </xsd:complexType>
</xsd:schema>

Save the above script as XX_FILENAME.xsd

STEP 2: BUILD OIC INTEGRATION

Create Application Based Integration which accepts the request body as Binay Format.

Read file using stage action as below

Remember to upload your XSD file here as below in the stage read file operation

STEP 3: Write the file for Notification

Using stage action with Write Operation , write a file which has been read in the early step
make sure upload the same XSD file for sample data.

STEP 4: SEND NOTIFICATIONS

Hope we are familiar with OIC notification as below map data from write operation.

Finally, we read & write the non-uniform csv file in OIC successfully.

Note: ORACLE DOC reference for XSD Script

Calling OIC Integration from Oracle Fusion via HTTPS

Naveen — Mon, 08 Jun 2026 02:37:20 +0000

Oracle Fusion Applications often require integration with external or internal services to extend business capabilities. Oracle Integration Cloud (OIC) serves as a powerful middleware platform to orchestrate such integrations.

One of the most common patterns is invoking an OIC integration from Fusion using an HTTPS connection, ensuring secure and reliable communication.

In this blog, we'll walk through the end-to-end process, including setup, configuration, and best practices

Step 1: Create an OIC Integration

Login to Oracle Integration Cloud
Navigate to Integrations → Create
Choose:REST Adapter (recommended for HTTPS calls)

Configure:

 * Resource URI (e.g., /customer/update)
 * HTTP Method (POST/GET)
 * Request/Response payload (JSON/XML)

Reference

Request Payload:

Make sure you selected
What is the media-type of Request Body? (Content-Type Header)
URL-encoded for https call acceptance

Response Payload:
I drop the xml reference here for response payload

<?xml version="1.0" encoding="UTF-8"?>
<xm:schema xmlns:xm="http://www.w3.org/2001/XMLSchema">
<!-- Root element-->
<xm:element name="root" type="root"/>
<!-- Complex type-->
    <xm:complexType name="root">
      <xm:sequence>
        <xm:element name="outcome" type="xm:float"/>
        <xm:element name="Instance_ID" type="xm:string"/>
              </xm:sequence>
            </xm:complexType>
</xm:schema>

Step 2: Create an JavaScript function using libraries

Navigations

Under Design -->libraries-->upload your js function

Step 3:Call Js function in integration

call Js function in your designing & map respective inputs to your function

Step 4: Map the response of integration

As below maps the final response.

Map the primary business identifiers
Active the Integration
Try to run & collect your endpoint Metadata

Step 5: Fusion BIP https connection configuration
Now we will map the integration connection details in fusion .

So for that we will login fusion/analytics-->Under profile Icon right side [Administration]-->Publisher[Manage Publisher]

Under HTTP creation data source with OIC login details

Step 6: Fusion BIP Data Model creation

Now choose the Data Source as HTTPS connection

fill respective details as below

Step 7: Testing

Run the data model with passing parameters , it will hit the integration as Input which we earlier given as request payload.
The response will display as below, which is back from integration.

Now we will see the Integration track Instance

By following the steps above setting up OIC, configuring HTTPS calls, handling authentication, and securing communication you can build reliable and scalable integrations.

Change VBCS Default Oracle Background to a Custom Image

Naveen — Sat, 06 Jun 2026 16:01:02 +0000

Oracle Visual Builder Cloud Service (VBCS) provides a clean default UI, but often, applications require branding or a more personalized look and feel. One simple yet effective customization is replacing the default background with a custom image.

In this blog, we'll walk through how to change the default Oracle VBCS background to a custom image in a few easy steps.

Step 1: Upload Your Custom Image

By default, applications have UI like below

Let's upload your custom image in VBCS application

Using Control & Navigation will upload image

Now we will upload image in data as below

Theme also choose here

Run Preview of your application page & Inspect image for entire VBCS Url

Step 2: Add Custom CSS

Make sure you will remove the image component from the page after you got the Url

Step 3: Final Preview

Note: Use optimized images (for faster loading)

Agentless Polling in Oracle Integration Using Public Gateway and ATP

Naveen — Fri, 05 Jun 2026 15:13:41 +0000

Enabling Secure Inbound Polling Without Connectivity Agent

In Oracle Integration (OIC), polling-based integrations are a common way to process database-driven events. This blog walks through a simple and practical polling use case using the classic EMP table, enhanced with a custom column to control processing.

Use Case Overview
We want to:

Add a column to track whether a record needs processing
Use OIC to poll records from EMP table
Process only specific rows
Update status after processing

Step 1: Create ATP Connection
The Below is Reference for Oracle ATP connection in OIC.

Step 2: Create Integration

I Prefer the Application Integration, Polling mechanism will work in the Initial Endpoint in the Integration.
Call ATP Here as like below.

Make sure choose the Review the polling strategy and specify polling options

Note: Apply your requirement handling in the read & unread value W.R.T perform on the Column [HIKE is a column for my case]

Whenever I update in HIKE column to "NO", The integration polling happens using ATP Polling Event & It will update the respective rows to "YES"

The Final Integration Design as below
I mapped EMPNAME as Primary Business identifier for track.

Step 3:TEST
I will update as below

Step 4:Integration Observation

It will automatically calls & update the field to "YES"

Full Flow Summary

DB Record Update → HIKE = 'NO'
OIC Polling Query Runs
Record is Picked
Integration Logic Executes
Record Updated → HIKE = 'YES'

Conclusion

This EMP table polling example demonstrates a lightweight and efficient pattern for OIC integrations:

Add a flag column
Poll using a filter condition
Process and update status

Combined with OIC Public Gateway and ATP Adapter, this becomes a fully cloud-native solution without any Connectivity Agent dependency.

OIC Developer Guide to Create AR Customer Contact Email & Phone in Oracle Fusion Using SOAP (mergePerson)

Naveen — Thu, 04 Jun 2026 16:01:53 +0000

In Oracle Fusion, AR Customer Contacts are stored as Party Relationships. Adding or updating contact details such as phone number and email address requires interaction with the Foundation Person Service SOAP API.
This guide walks you through:

Identifying required data (PartyId & RelationshipId)
Using SOAP API to create contact points
Verifying data in backend tables

Step 1: Identify Customer Contact Details
Relationship Information

select * from fusion.hz_relationships where relationship_id=300000159607279

Party Information

select * from fusion.hz_parties where party_id= 300000159607277

Step 2: Use SOAP API to Create Contact Details
WSDL Endpoint:
Use SOAP API /crmService/FoundationPartiesPersonService to add new phone number and email for this AR Customer Contact

Operation:mergePerson
This API is used to update or merge person-related information, including contact points.

Step 3: Sample SOAP Request
Below is a working payload to create both email and phone:

<soapenv:Envelope xmlns:soapenv="http://schemas.xmlsoap.org/soap/envelope/"
                  xmlns:typ="http://xmlns.oracle.com/apps/cdm/foundation/parties/personService/applicationModule/types/"
                  xmlns:per="http://xmlns.oracle.com/apps/cdm/foundation/parties/personService/"
                  xmlns:rel="http://xmlns.oracle.com/apps/cdm/foundation/parties/relationshipService/"
                  xmlns:con="http://xmlns.oracle.com/apps/cdm/foundation/parties/contactPointService/">
   <soapenv:Header/>
   <soapenv:Body>
      <typ:mergePerson>
         <typ:personParty>
            <per:PartyId>300000159607277</per:PartyId>

            <per:Relationship>
               <rel:RelationshipId>300000159607279</rel:RelationshipId>

               <!-- Email Creation -->
               <rel:Email>
                  <con:OwnerTableName>HZ_PARTIES</con:OwnerTableName>
                  <con:CreatedByModule>HZ_WS</con:CreatedByModule>
                  <con:ContactPointPurpose>BUSINESS</con:ContactPointPurpose>
                  <con:EmailAddress>Addtest@test.com</con:EmailAddress>
               </rel:Email>

               <!-- Phone Creation -->
               <rel:Phone>
                  <con:OwnerTableName>HZ_PARTIES</con:OwnerTableName>
                  <con:CreatedByModule>HZ_WS</con:CreatedByModule>
                  <con:ContactPointPurpose>BUSINESS</con:ContactPointPurpose>
                  <con:PhoneNumber>2128811223</con:PhoneNumber>
                  <con:PhoneExtension>999</con:PhoneExtension>
                  <con:PhoneLineType>MOBILE</con:PhoneLineType>
               </rel:Phone>

            </per:Relationship>
         </typ:personParty>
      </typ:mergePerson>
   </soapenv:Body>
</soapenv:Envelope>

Step 4: Validate Data in Frontend Screen

Step 5: Validate Data in Backend Tables
After successful API execution, verify using:

SELECT CONTACT_POINT_ID,
       CONTACT_POINT_TYPE,
       PHONE_TYPE,
       OWNER_TABLE_ID,
       RELATIONSHIP_ID,
       PHONE_NUMBER,
       EMAIL_ADDRESS
FROM fusion.hz_contact_points
WHERE creation_date > SYSDATE - 1
  AND created_by_module = 'HZ_WS'
  AND owner_table_name = 'HZ_PARTIES';

Step 6: From OIC

With same WSDL :https:/crmService/FoundationPartiesPersonService?WSDL
create Connection.
Call the connection in OIC with mergePerson Operation
Pass respective Values to it

Conclusion:
We successfully updated the Customer communication details using SOAP WSDL in Oracle Fusion ♥

DEV Community: Naveen

AI-Powered BI Publisher Report Retrieval in Oracle Fusion Agent AI 26C

Project Vision

Example Queries

Solution Architecture

Agent Role:

Prompt:

Create External Tool

Request Body

Challenge Encountered

One integration, N partners — routed at runtime by dynamic connections

Let's DO it Dynamic Connections in OIC & Build Multi-tenant partner router Project

The core idea

How the resolution mechanism actually works

The project: a multi-tenant partner router

The routing table(Lookup)

The flow

Lessons from actually building and testing this

Where else this pattern applies

Building a Real-Time Collaborative Notes App on Oracle Backend with Firebase APIs

The project

The Oracle technology behind it

1. Real-time sync via onSnapshot

2. File uploads with inline preview

3. A collaborative permission model, enforced server-side

What users actually get out of this

How it's deployed

Try it / read the code

Using PL/SQL in Oracle Fusion BI Publisher Reports for Dynamic Data Filtering

Create Data model in Fusion

Real-Time Reporting: How to Push ATP Data to Oracle Fusion BI Publisher

Automating Customer Termination Lifecycle Using Oracle Integration Cloud, ATP, and Oracle Service Cloud

Introduction

Solution Architecture

Integration Flow

Business Value Delivered

Stop Using Q! Use Finder for Faster Oracle OIC REST Integrations

STRUGGLING WITH IRREGULAR CSV FILES IN OIC? HERE’S THE NXSD SOLUTION

Calling OIC Integration from Oracle Fusion via HTTPS

Change VBCS Default Oracle Background to a Custom Image

Agentless Polling in Oracle Integration Using Public Gateway and ATP

Enabling Secure Inbound Polling Without Connectivity Agent

OIC Developer Guide to Create AR Customer Contact Email & Phone in Oracle Fusion Using SOAP (mergePerson)

1. Real-time sync via `onSnapshot`