Building AI-Powered Image Similarity Search in Oracle APEX using Cohere & Oracle 23ai Vector


🧠 Introduction

In this blog, we’ll build a real-time Image Similarity Search Application using:

  • Oracle APEX (UI)

  • Oracle 23ai VECTOR (AI storage)

  • Cohere Embed API (Image → Vector)

  • ORDS (Image streaming)

Instead of searching by text, users can upload an image and find visually similar images instantly.

πŸ—️ Architecture Overview

The system works in 5 simple steps:

  1. Upload image in APEX

  2. Convert image → vector using Cohere API

  3. Store vector in Oracle DB

  4. Compare vectors using cosine similarity

  5. Display matching images in classic report.

Step by Step Approaches 

πŸ—„️ Step 1: Create Database Tables

 CREATE TABLE "PRODUCT_IMAGES" 

   ( "ID" NUMBER GENERATED ALWAYS AS IDENTITY, 
"PRODUCT_NAME" VARCHAR2(200), 
 "IMAGE_BLOB" BLOB, 
 "IMAGE_MIME" VARCHAR2(100), 
 "EMBEDDING" VECTOR (1536, FLOAT32), 
 PRIMARY KEY ("ID")
  USING INDEX ENABLE
   );

CREATE TABLE "SEARCH_IMAGE" 

   ( "ID" NUMBER GENERATED ALWAYS AS IDENTITY, 

"PRODUCT_NAME" VARCHAR2(200), 

"IMAGE_BLOB" BLOB, 

"IMAGE_MIME" VARCHAR2(100), 

"EMBEDDING" VECTOR (1536, FLOAT32), 

PRIMARY KEY ("ID")

  USING INDEX ENABLE

   );

πŸ‘‰ VECTOR(1536) matches Cohere embed-v4.0 output.

πŸ”‘ Step 2: Get Cohere API Key

  • Go to Cohere Dashboard - https://dashboard.cohere.com/

  • Click API Keys option and Create API Key (Trial is enough)

πŸ–Ό️ Step 3: Upload & Store Product Image Embeddings

πŸ‘‰ This runs when uploading product images

Key Logic:

  • Read image from APEX temp table

  • Convert to Base64

  • Call Cohere API

  • Store embedding in DB

Steps:

Create a new page in your Oracle APEX application and add the following page items:

  • P2_PRODUCT_NAME – Text field to enter the product name

  • P_UPLOAD_IMAGE – File Browse item to upload the product image

  • P_STATUS – Display Only item to show success or error messages

Next, create a button named UPLOAD_IMAGE to trigger the upload process.

Finally, create a page process called Insert Image, which will handle storing the uploaded image along with its embedding into the database table.




Process code:

DECLARE
    l_mime         VARCHAR2(100);
    l_blob         BLOB;
    l_blob_req     BLOB;
    l_base64       CLOB;
    l_json_req     CLOB;
    l_json_resp    CLOB;
    l_raw_chunk    RAW(7500);
    l_b64_chunk    VARCHAR2(10000);
    l_offset       INTEGER := 1;
    l_chunk_size   INTEGER := 7500;
    l_amount       INTEGER;
    l_blob_len     INTEGER;
    l_dest_off     INTEGER := 1;
    l_src_off      INTEGER := 1;
    l_lang_ctx     INTEGER := DBMS_LOB.DEFAULT_LANG_CTX;
    l_warning      INTEGER;
    l_vector_str   CLOB;
    l_dim_count    INTEGER;
    l_val          VARCHAR2(50);
BEGIN
    -- Step 1: Fetch image
    SELECT blob_content, mime_type
    INTO   l_blob, l_mime
    FROM   (
        SELECT blob_content, mime_type
        FROM   APEX_APPLICATION_TEMP_FILES
        WHERE  name = :P_UPLOAD_IMAGE
        ORDER  BY created_on DESC
    )
    WHERE ROWNUM = 1;
IF l_mime LIKE 'image/jpg%' OR l_mime = 'image/pjpeg' THEN
    l_mime := 'image/jpeg';
END IF;
    l_blob_len := DBMS_LOB.GETLENGTH(l_blob);
    IF l_blob_len IS NULL OR l_blob_len = 0 THEN
        RAISE_APPLICATION_ERROR(-20001, 'Image BLOB is empty or NULL');
    END IF;
 
    -- Step 2: BLOB → Base64 CLOB (no newlines)
    DBMS_LOB.CREATETEMPORARY(l_base64, TRUE);
    l_base64 := APEX_WEB_SERVICE.BLOB2CLOBBASE64(l_blob);
     l_base64 := REPLACE(REPLACE(l_base64, CHR(10), ''), CHR(13), '');
    -- Step 3: Build JSON WITH data URI prefix (required by Cohere v2)
DBMS_LOB.CREATETEMPORARY(l_json_req, TRUE);
DBMS_LOB.APPEND(l_json_req, TO_CLOB('{"model":"embed-v4.0",'));
DBMS_LOB.APPEND(l_json_req, TO_CLOB('"input_type":"image",'));
DBMS_LOB.APPEND(l_json_req, TO_CLOB('"embedding_types":["float"],'));
DBMS_LOB.APPEND(l_json_req, TO_CLOB('"images":["'));
DBMS_LOB.APPEND(l_json_req, TO_CLOB('data:' || l_mime || ';base64,'));
DBMS_LOB.APPEND(l_json_req, l_base64);
DBMS_LOB.APPEND(l_json_req, TO_CLOB('"]}'));
 
    -- Step 4: Convert JSON CLOB → BLOB (avoids 32767 limit)
    DBMS_LOB.CREATETEMPORARY(l_blob_req, TRUE);
    DBMS_LOB.CONVERTTOBLOB(
        dest_lob     => l_blob_req,
        src_clob     => l_json_req,
        amount       => DBMS_LOB.LOBMAXSIZE,
        dest_offset  => l_dest_off,
        src_offset   => l_src_off,
        blob_csid    => DBMS_LOB.DEFAULT_CSID,
        lang_context => l_lang_ctx,
        warning      => l_warning
    );
 
    -- Step 5: Call Cohere API
    APEX_WEB_SERVICE.G_REQUEST_HEADERS.DELETE;
    APEX_WEB_SERVICE.G_REQUEST_HEADERS(1).NAME  := 'Content-Type';
    APEX_WEB_SERVICE.G_REQUEST_HEADERS(1).VALUE := 'application/json';
    APEX_WEB_SERVICE.G_REQUEST_HEADERS(2).NAME  := 'Authorization';
    APEX_WEB_SERVICE.G_REQUEST_HEADERS(2).VALUE := 'Bearer YuJE9V6b9FiWzpmyIYtTh7IiGuidOCd2I0m3YBca';
    l_json_resp := APEX_WEB_SERVICE.MAKE_REST_REQUEST
    (
        p_url         => 'https://api.cohere.com/v2/embed',
        p_http_method => 'POST',
        p_body_blob   => l_blob_req
    );
 
    -- Step 6: Show result
    IF APEX_WEB_SERVICE.G_STATUS_CODE = 200 THEN
        :P_STATUS := 'success';
    APEX_JSON.PARSE(l_json_resp);
 
        -- Cohere v2 response path: embeddings.float[1][n]
        l_dim_count := APEX_JSON.GET_COUNT(p_path => 'embeddings.float[1]');
        IF l_dim_count IS NULL OR l_dim_count = 0 THEN
            RAISE_APPLICATION_ERROR(-20002, 'No embedding dimensions found in response');
        END IF;
 
        -- Build vector string: '[v1,v2,...,v1024]'
        DBMS_LOB.CREATETEMPORARY(l_vector_str, TRUE);
        DBMS_LOB.APPEND(l_vector_str, TO_CLOB('['));
        FOR i IN 1 .. l_dim_count LOOP
            l_val := TO_CHAR(
                         APEX_JSON.GET_NUMBER(p_path => 'embeddings.float[1][%d]', p0 => i),
                         'FM999999990.9999999999'
                     );
            IF i > 1 THEN
                DBMS_LOB.APPEND(l_vector_str, TO_CLOB(','));
            END IF;
            DBMS_LOB.APPEND(l_vector_str, TO_CLOB(l_val));
        END LOOP;
        DBMS_LOB.APPEND(l_vector_str, TO_CLOB(']'));
 
        -- Insert into PRODUCT_IMAGES
        INSERT INTO PRODUCT_IMAGES (
            product_name,
            image_blob,
            image_mime,
            embedding
        ) VALUES (
            :P2_PRODUCT_NAME,                          -- bind your product name item
            l_blob,
            l_mime,
            TO_VECTOR(l_vector_str, 1536, FLOAT32)     -- Oracle 23ai VECTOR constructor
        );
        COMMIT;
 
        --:P2_QUERY_VECTOR := 'SUCCESS — Stored ' || l_dim_count || ' dimensions, ID: ';
 
        DBMS_LOB.FREETEMPORARY(l_vector_str);
    else
     :P_STATUS := 'HTTP ' || APEX_WEB_SERVICE.G_STATUS_CODE|| ' | ' || SUBSTR(l_json_resp, 1,3000);
    END IF;
 
    -- Step 7: Cleanup
    DBMS_LOB.FREETEMPORARY(l_base64);
    DBMS_LOB.FREETEMPORARY(l_json_req);
    DBMS_LOB.FREETEMPORARY(l_blob_req);
 
EXCEPTION
    WHEN OTHERS THEN
        :P_STATUS  := 'ERROR [' || SQLCODE || ']: ' || SQLERRM
                            || ' | ' || DBMS_UTILITY.FORMAT_ERROR_BACKTRACE();
END;

πŸ” Step 4: Store Search Image Embedding 

What it does:

  • Reads uploaded image

  • Converts to Base64

  • Calls Cohere API

  • Extracts 1536 vector values

  • Stores in SEARCH_IMAGE

Steps:

Create a new page in your Oracle APEX application and add the following page items:

  • P4_UPLOAD_IMAGE – File Browse item to upload the search image

  • P4_UPLOADED_IMAGE – Display item to preview the uploaded image

  • P_SEARCH_RESULTS – Hidden item to store the JSON response of similar images

  • P4_STATUS – Display Only item to show success or error messages

Next, create a button named SEARCH_SIMILAR_IMAGE to trigger the search process.

Add a Classic Report region to display the similar images retrieved from the database.

Finally, create two page processes:

  1. Insert Image Process – Converts the uploaded image into a vector using the Cohere API and stores it in the SEARCH_IMAGE table

  2. Search Similar Images Process – Compares the uploaded image vector with existing product images and returns the top matching results



Classic Report Code:

SELECT
    ranked.rank#,
    ranked.product_id                              AS id,
    ranked.product_name,
    ranked.similarity                              AS similarity_pct,
    '<div style="text-align:center;">'
    || '<img src="'
    || 'https://oracleapex.com/ords/sabarinath/product-images/'
    || ranked.product_id
    || '" style="width:300px;height:200px;object-fit:scale-down;'
    || 'border-radius:8px;border:1px solid #ddd;" />'
    || '<br/><b>' || ranked.product_name || '</b>'
    || '<br/>Match: <span style="color:green;font-weight:bold;">'
    || ranked.similarity || '%</span>'
    || '</div>'                                    AS image_html
FROM (
    SELECT
        ROWNUM                        AS rank#,
        TO_NUMBER(jt.product_id)      AS product_id,
        jt.product_name,
        TO_NUMBER(jt.similarity)      AS similarity
    FROM
        JSON_TABLE(
            :P_SEARCH_RESULTS,
            '$[*]'
            COLUMNS (
                product_id   VARCHAR2(20)  PATH '$.product_id',
                product_name VARCHAR2(200) PATH '$.product_name',
                mime_type    VARCHAR2(100) PATH '$.mime_type',
                similarity   VARCHAR2(20)  PATH '$.similarity'
            )
        ) jt
    ORDER BY TO_NUMBER(jt.similarity) DESC
) ranked
JOIN PRODUCT_IMAGES pi
    ON pi.id = ranked.product_id
WHERE ranked.similarity > 60

Insert image Process Code:

DECLARE
    l_mime         VARCHAR2(100);
    l_blob         BLOB;
    l_blob_req     BLOB;
    l_base64       CLOB;
    l_json_req     CLOB;
    l_json_resp    CLOB;
    l_raw_chunk    RAW(7500);
    l_b64_chunk    VARCHAR2(10000);
    l_offset       INTEGER := 1;
    l_chunk_size   INTEGER := 7500;
    l_amount       INTEGER;
    l_blob_len     INTEGER;
    l_dest_off     INTEGER := 1;
    l_src_off      INTEGER := 1;
    l_lang_ctx     INTEGER := DBMS_LOB.DEFAULT_LANG_CTX;
    l_warning      INTEGER;
    l_vector_str   CLOB;
    l_dim_count    INTEGER;
    l_val          VARCHAR2(50);
BEGIN
    -- Step 1: Fetch image
    SELECT blob_content, mime_type
    INTO   l_blob, l_mime
    FROM   (
        SELECT blob_content, mime_type
        FROM   APEX_APPLICATION_TEMP_FILES
        WHERE  name = :P4_UPLOAD_IMAGE
        ORDER  BY created_on DESC
    )
    WHERE ROWNUM = 1;
IF l_mime LIKE 'image/jpg%' OR l_mime = 'image/pjpeg' THEN
    l_mime := 'image/jpeg';
END IF;
    l_blob_len := DBMS_LOB.GETLENGTH(l_blob);
    IF l_blob_len IS NULL OR l_blob_len = 0 THEN
        RAISE_APPLICATION_ERROR(-20001, 'Image BLOB is empty or NULL');
    END IF;
 
    -- Step 2: BLOB → Base64 CLOB (no newlines)
    DBMS_LOB.CREATETEMPORARY(l_base64, TRUE);
    l_base64 := APEX_WEB_SERVICE.BLOB2CLOBBASE64(l_blob);
     l_base64 := REPLACE(REPLACE(l_base64, CHR(10), ''), CHR(13), '');
    -- Step 3: Build JSON WITH data URI prefix (required by Cohere v2)
DBMS_LOB.CREATETEMPORARY(l_json_req, TRUE);
DBMS_LOB.APPEND(l_json_req, TO_CLOB('{"model":"embed-v4.0",'));
DBMS_LOB.APPEND(l_json_req, TO_CLOB('"input_type":"image",'));
DBMS_LOB.APPEND(l_json_req, TO_CLOB('"embedding_types":["float"],'));
DBMS_LOB.APPEND(l_json_req, TO_CLOB('"images":["'));
DBMS_LOB.APPEND(l_json_req, TO_CLOB('data:' || l_mime || ';base64,'));
DBMS_LOB.APPEND(l_json_req, l_base64);
DBMS_LOB.APPEND(l_json_req, TO_CLOB('"]}'));
 
    -- Step 4: Convert JSON CLOB → BLOB (avoids 32767 limit)
    DBMS_LOB.CREATETEMPORARY(l_blob_req, TRUE);
    DBMS_LOB.CONVERTTOBLOB(
        dest_lob     => l_blob_req,
        src_clob     => l_json_req,
        amount       => DBMS_LOB.LOBMAXSIZE,
        dest_offset  => l_dest_off,
        src_offset   => l_src_off,
        blob_csid    => DBMS_LOB.DEFAULT_CSID,
        lang_context => l_lang_ctx,
        warning      => l_warning
    );
 
    -- Step 5: Call Cohere API
    APEX_WEB_SERVICE.G_REQUEST_HEADERS.DELETE;
    APEX_WEB_SERVICE.G_REQUEST_HEADERS(1).NAME  := 'Content-Type';
    APEX_WEB_SERVICE.G_REQUEST_HEADERS(1).VALUE := 'application/json';
    APEX_WEB_SERVICE.G_REQUEST_HEADERS(2).NAME  := 'Authorization';
    APEX_WEB_SERVICE.G_REQUEST_HEADERS(2).VALUE := 'Bearer YuJE9V6b9FiWzpmyIYtTh7IiGuidOCd2I0m3YBca';
 
    l_json_resp := APEX_WEB_SERVICE.MAKE_REST_REQUEST
    (
        p_url         => 'https://api.cohere.com/v2/embed',
        p_http_method => 'POST',
        p_body_blob   => l_blob_req
    );
 
    -- Step 6: Show result
    IF APEX_WEB_SERVICE.G_STATUS_CODE = 200 THEN
        :P_STATUS := 'success';
    APEX_JSON.PARSE(l_json_resp);
 
        -- Cohere v2 response path: embeddings.float[1][n]
        l_dim_count := APEX_JSON.GET_COUNT(p_path => 'embeddings.float[1]');
 
        IF l_dim_count IS NULL OR l_dim_count = 0 THEN
            RAISE_APPLICATION_ERROR(-20002, 'No embedding dimensions found in response');
        END IF;
 
        -- Build vector string: '[v1,v2,...,v1024]'
        DBMS_LOB.CREATETEMPORARY(l_vector_str, TRUE);
        DBMS_LOB.APPEND(l_vector_str, TO_CLOB('['));
 
        FOR i IN 1 .. l_dim_count LOOP
            l_val := TO_CHAR(
                         APEX_JSON.GET_NUMBER(p_path => 'embeddings.float[1][%d]', p0 => i),
                         'FM999999990.9999999999'
                     );
            IF i > 1 THEN
                DBMS_LOB.APPEND(l_vector_str, TO_CLOB(','));
            END IF;
            DBMS_LOB.APPEND(l_vector_str, TO_CLOB(l_val));
        END LOOP;
        delete from search_image;
        DBMS_LOB.APPEND(l_vector_str, TO_CLOB(']'));
 
        -- Insert into PRODUCT_IMAGES
        INSERT INTO SEARCH_IMAGE (
            product_name,
            image_blob,
            image_mime,
            embedding
        ) VALUES (
            'Search_image',                          -- bind your product name item
            l_blob,
            l_mime,
            TO_VECTOR(l_vector_str, 1536, FLOAT32)     -- Oracle 23ai VECTOR constructor
        );
 
        COMMIT;
 
        --:P7_QUERY_VECTOR := 'SUCCESS — Stored ' || l_dim_count || ' dimensions, ID: ';
 
        DBMS_LOB.FREETEMPORARY(l_vector_str);
 
    else
 
     :P4_STATUS := 'HTTP ' || APEX_WEB_SERVICE.G_STATUS_CODE|| ' | ' || SUBSTR(l_json_resp, 1,3000);
 
        
    END IF;
 
    -- Step 7: Cleanup
    DBMS_LOB.FREETEMPORARY(l_base64);
    DBMS_LOB.FREETEMPORARY(l_json_req);
    DBMS_LOB.FREETEMPORARY(l_blob_req);
 
EXCEPTION
    WHEN OTHERS THEN
        :P4_STATUS  := 'ERROR [' || SQLCODE || ']: ' || SQLERRM
                            || ' | ' || DBMS_UTILITY.FORMAT_ERROR_BACKTRACE();
END;

πŸ”Ž Step 5: Search Similar Images (Vector Comparison)

 This is your second process in same page.

Search Similiar Images Process Code:

DECLARE
    l_search_vector    VECTOR(1536, FLOAT32);
    l_result_json      CLOB;
BEGIN
    -- Step 1: Get the latest (or specific) search image vector
    SELECT embedding
    INTO   l_search_vector
    FROM   (
        SELECT embedding
        FROM   SEARCH_IMAGE
        ORDER  BY id DESC
    )
    WHERE ROWNUM = 1;
 
    -- Step 2: Build JSON result of top-N matches (for APEX display)
    SELECT JSON_ARRAYAGG(
               JSON_OBJECT(
                   'product_id'   VALUE p.id,
                   'product_name' VALUE p.product_name,
                   'mime_type'    VALUE p.image_mime,
                   'similarity'   VALUE ROUND(
                                      (1 - VECTOR_DISTANCE(p.embedding, l_search_vector, COSINE)) * 100,
                                     2
                                  )
               )
               ORDER BY VECTOR_DISTANCE(p.embedding, l_search_vector, COSINE) ASC
           )
    INTO  l_result_json
    FROM  PRODUCT_IMAGES p
    WHERE p.embedding IS NOT NULL
      AND ROWNUM <= 20;   -- top 5 matches
 
    -- Step 3: Bind to APEX page item (e.g. a Hidden or Display item)
    :P_SEARCH_RESULTS := l_result_json;
    :P4_STATUS         := 'success';
EXCEPTION
    WHEN NO_DATA_FOUND THEN
        :P4_STATUS := 'No search vector found.';
    WHEN OTHERS THEN
        :P4_STATUS := 'ERROR [' || SQLCODE || ']: ' || SQLERRM
                     || ' | ' || DBMS_UTILITY.FORMAT_ERROR_BACKTRACE();
END;

Logic:

  • Get latest search vector

  • Compare with all product vectors

  • Return top matches as JSON

πŸ‘‰ Output stored in:

:P_SEARCH_RESULTS

πŸ“Š Step 6: Display Results (Classic Report)

Use this concept:

  • Parse JSON using JSON_TABLE

  • Show:

    • Image

    • Product Name

    • Similarity %

Similarity Formula:

Similarity % = (1 - cosine distance) * 100

🌐 Step 7: ORDS Image Rendering (Step-by-Step)
To display images without session issues in Oracle APEX, we use ORDS REST Services to stream
images directly from the database.
πŸ”§ Step 7.1: Create REST Module
  1. Go to
SQL Workshop → RESTful Services
  2. Click Modules → Create Module
  3. Enter:
  • Module Name: product-images
  • Base Path: /product-images/
  • Status: Published
πŸ‘‰ Click Create Module
πŸ”§ Step 7.2: Create Resource Template
  1. Open your module: product-images
  2. Click Create Template
  3. Enter:
  • URI Template: :id
πŸ‘‰ Click Create Template
πŸ”§ Step 7.3: Create Resource Handler
  1. Inside the template, click Create Handler
  2. Configure:
  • Method: GET
  • Source Type: Media Resource
  • Format: Binary Representation
🧾 Step 7.4: Add SQL Query (Handler Source)
Paste the below SQL:
SELECT image_mime AS content_type,
       image_blob AS content
FROM PRODUCT_IMAGES
WHERE id =:id
πŸ‘‰ This tells ORDS:
  • content_type → MIME type (image/jpeg, image/png)
  • content → actual image BLOB
πŸ‘‰ This avoids:
  • Session issues
  • Broken images
  • Performance problems
✅ Step 7.5: Apply Changes
  • Click Apply Changes
  • Ensure module is Published
🌍 Step 7.6: Test URL
Use this format:
https://oracleapex.com/ords/{workspace}/product-images/{id}
Example:
https://oracleapex.com/ords/sabarinath/product-images/1
πŸ‘‰ It should directly display the image in browser
⚙️ Step 8: APEX Page Configuration
Page 2 (Product Upload)
  • Product Name
  • File Upload
  • Process → Store embedding
Page 4 (Search Image)
  • File Upload
  • Button: Search Similar Image
Processes:
  1. Insert Image (Embedding)
  2. Search Similar Images
  3. Refresh Report
🎯 Step 9: Final Output
User flow:
  1. Upload image
  2. Click Search Similar Image
  3. See results like:
  • Car → Similar Cars
  • Bike → Similar Bikes


πŸš€ Advanced Improvements
You can enhance this project by adding:
  • πŸ” Similarity slider (dynamic threshold)
  • πŸ“‚ Category-based filtering
  • ⚡ Vector Index for performance
  • πŸ€– Multi-modal search (text + image)
🏁 Conclusion
We successfully built an AI-powered Image Similarity Search system in Oracle APEX using:
  • Cohere for embeddings
  • Oracle 23ai VECTOR for storage
  • APEX for UI
This is a powerful real-world use case of AI inside Oracle applications.

Comments

Post a Comment

Popular posts from this blog

Building Secure RESTful Services in Oracle APEX 24.2 Using ORDS, OAuth2 Client Credentials, and PL/SQL

Image
πŸ“˜ Introduction Oracle APEX 24.2 provides powerful capabilities to expose database data as secure RESTful APIs using Oracle REST Data Services (ORDS). In modern enterprise architectures, APIs are essential for enabling system-to-system communication. Instead of direct database access, RESTful services allow controlled and secure data sharing. In this blog, we will walk through how to create a REST API in Oracle APEX 24.2 and secure it using privileges and OAuth authentication. 🎯 Objective The goal of this implementation is to: • Expose database data through REST APIs • Create structured API endpoints • Secure APIs using OAuth • Enable controlled external access 🧱 Prerequisites Before starting, ensure: • Oracle APEX 24.2 is installed • ORDS 20.x or later is configured • RESTful Services are enabled • A working table exists (Example: HR.EMPLOYEES) 🌐 REST API URL Structure The generated API follows this format: https://host:port/ords/schema/module/template ⚙️ Implementat...
Read more

Building a Portfolio Generator in Oracle APEX: A Step-by-Step Guide to Generate Downloadable Portfolio Documents

Image
  Introduction This tutorial demonstrates how to build a Portfolio Generator application in Oracle APEX where users can enter resume details and generate a downloadable portfolio HTML/ PDF with one click . The application collects user details, work experience, skills, and education, then dynamically generates a styled HTML portfolio that can be printed as PDF . Step- by- Step  Implementation  Guide Step 1 -  Create Database Tables First create the tables required to store resume information. We will create 5 tables . 1️⃣ Resume Master Table Stores the main profile information.   CREATE TABLE "RESUME_MASTER"     ( "RESUME_ID" NUMBER GENERATED ALWAYS AS IDENTITY, "FULL_NAME" VARCHAR2(200),  "EMAIL" VARCHAR2(200),  "PHONE" VARCHAR2(50),  "ADDRESS" VARCHAR2(500),  "DOB" DATE,  "OBJECTIVE" CLOB,  "CURRENT_COMPANY" VARCHAR2(200),  "CURRENT_ROLE" VARCHAR2(200),  "C...
Read more

Implementing WhatsApp OTP Verification in Oracle APEX Using UltraMsg API

Image
  πŸš€ Introduction In today’s digital applications, user verification plays a crucial role in ensuring security and trust. Traditionally, OTPs are sent via SMS — but with the widespread use of WhatsApp, sending OTPs through WhatsApp offers: Faster delivery Better user reach Improved engagement In this blog, we explore how to implement OTP verification in Oracle APEX by integrating WhatsApp messaging using an external API service. 🎯 Objective The goal of this implementation was to: ✔️ Generate OTP dynamically ✔️ Send OTP via WhatsApp ✔️ Integrate external messaging API ✔️ Enhance login/verification security ✔️ Enable real-time communication 🧩 Solution Overview This solution integrates: PL/SQL OTP generation REST API call from Oracle APEX WhatsApp messaging service This allows Oracle APEX applications to send OTP directly to users via WhatsApp instead of SMS.  πŸ§± Step-by-Step Approach  Step 1: Configure WhatsApp API To send OTP via WhatsApp, ...
Read more