new operation added

read scanned documents (pdf only)
PDF will be split into images by page and send to image model to convert to the request in the prompt.

pom.xml

@@ -5,7 +5,7 @@
 	<modelVersion>4.0.0</modelVersion>
 	<groupId>com.mule.mulechain</groupId>
 	<artifactId>mulechain-ai-connector</artifactId>
-	<version>0.2.6</version>
+	<version>0.2.10</version>
 	<packaging>mule-extension</packaging>
 	<name>MuleChain</name>
 
@@ -168,6 +168,11 @@
 			<artifactId>rxjava</artifactId>
 			<version>2.2.21</version>
 		</dependency>
+		<dependency>
+			<groupId>org.apache.pdfbox</groupId>
+			<artifactId>pdfbox</artifactId>
+			<version>3.0.2</version>
+		</dependency>
 	</dependencies>
 	<repositories>
 		<repository>

src/main/java/org/mule/extension/mulechain/internal/constants/MuleChainConstants.java

@@ -40,6 +40,9 @@ private MuleChainConstants() {}
   public static final String DATASET = "dataset";
   public static final String SENTIMENT = "sentiment";
   public static final String IS_POSITIVE = "isPositive";
+  public static final String PAGE = "page";
+  public static final String TOTAL_PAGES = "total_pages";
+  public static final String PAGES = "pages";
 
   public static class EmbeddingConstants {
 

src/main/java/org/mule/extension/mulechain/internal/operation/LangchainImageModelsOperations.java

@@ -3,6 +3,7 @@
  */
 package org.mule.extension.mulechain.internal.operation;
 
+import org.json.JSONArray;
 import org.json.JSONObject;
 import org.mule.extension.mulechain.internal.config.LangchainLLMConfiguration;
 import org.mule.extension.mulechain.internal.constants.MuleChainConstants;
@@ -26,6 +27,19 @@
 import org.slf4j.Logger;
 import org.slf4j.LoggerFactory;
 
+import java.awt.image.BufferedImage;
+import java.io.ByteArrayOutputStream;
+import java.io.File;
+import java.io.IOException;
+import java.util.Base64;
+import java.util.List;
+
+import javax.imageio.ImageIO;
+
+import org.apache.pdfbox.pdmodel.PDDocument;
+import org.apache.pdfbox.pdmodel.PDPage;
+import org.apache.pdfbox.rendering.PDFRenderer;
+import org.apache.pdfbox.Loader;
 
 /**
  * This class is a container for operations, every public method in this class will be taken as an extension operation.
@@ -76,4 +90,67 @@ public String drawImage(@Config LangchainLLMConfiguration configuration, String
 
     return jsonObject.toString();
   }
+
+  /**
+   * Reads an scanned document.
+   */
+  @MediaType(value = ANY, strict = false)
+  @Alias("IMAGE-read-scanned-documents")
+  public String readScannedDocumentPDF(@Config LangchainLLMConfiguration configuration, String data, String filePath) {
+
+    ChatLanguageModel model = configuration.getModel();
+
+    String sourceDir = filePath;
+
+    JSONObject jsonObject = new JSONObject();
+    JSONArray docPages = new JSONArray();
+    try (PDDocument document = Loader.loadPDF(new File(sourceDir))) {
+
+      PDFRenderer pdfRenderer = new PDFRenderer(document);
+      int totalPages = document.getNumberOfPages();
+      LOGGER.info("Total files to be converted -> " + totalPages);
+      jsonObject.put(MuleChainConstants.TOTAL_PAGES, totalPages);
+
+      JSONObject docPage;
+      for (int pageNumber = 0; pageNumber < totalPages; pageNumber++) {
+
+        BufferedImage image = pdfRenderer.renderImageWithDPI(pageNumber, 300);
+        LOGGER.info("Reading page -> " + pageNumber);
+
+        String imageBase64 = convertToBase64String(image);
+        UserMessage userMessage = UserMessage.from(
+                                                   TextContent.from(data),
+                                                   ImageContent.from(imageBase64, "image/png"));
+
+        Response<AiMessage> response = model.generate(userMessage);
+
+        docPage = new JSONObject();
+        docPage.put(MuleChainConstants.PAGE, pageNumber + 1);
+        docPage.put(MuleChainConstants.RESPONSE, response.content().text());
+        docPage.put(MuleChainConstants.TOKEN_USAGE, JsonUtils.getTokenUsage(response));
+        docPages.put(docPage);
+      }
+
+    } catch (IOException e) {
+      LOGGER.info("Error occurred while processing the file: " + e.getMessage());
+    }
+
+    jsonObject.put(MuleChainConstants.PAGES, docPages);
+
+    return jsonObject.toString();
+  }
+
+  private String convertToBase64String(BufferedImage image) {
+    String base64String;
+    try (ByteArrayOutputStream outputStream = new ByteArrayOutputStream()) {
+      ImageIO.write(image, "png", outputStream);
+      byte[] imageBytes = outputStream.toByteArray();
+      base64String = Base64.getEncoder().encodeToString(imageBytes);
+      return base64String;
+    } catch (IOException e) {
+      e.printStackTrace();
+      LOGGER.info("Error occurred while processing the file: " + e.getMessage());
+      return "Error";
+    }
+  }
 }