again with the PDF interfaces

src/main/java/com/mindee/image/ImageExtractor.java

@@ -4,15 +4,17 @@
 import com.mindee.geometry.PositionDataField;
 import com.mindee.input.InputSourceUtils;
 import com.mindee.input.LocalInputSource;
-import com.mindee.pdf.PDFBoxApi;
-import com.mindee.pdf.PDFOperation;
-import com.mindee.pdf.PdfPageImage;
 import java.awt.image.BufferedImage;
 import java.io.ByteArrayInputStream;
 import java.io.IOException;
 import java.util.ArrayList;
 import java.util.List;
 import javax.imageio.ImageIO;
+import org.apache.pdfbox.Loader;
+import org.apache.pdfbox.pdmodel.PDDocument;
+import org.apache.pdfbox.pdmodel.common.PDRectangle;
+import org.apache.pdfbox.rendering.ImageType;
+import org.apache.pdfbox.rendering.PDFRenderer;
 
 /**
  * Extract sub-images from an image.
@@ -22,14 +24,14 @@ public class ImageExtractor {
   private final String filename;
   private final String saveFormat;
 
-  public ImageExtractor(LocalInputSource source, PDFOperation pdfOperation) throws IOException {
+  public ImageExtractor(LocalInputSource source) throws IOException {
     this.filename = source.getFilename();
     this.pageImages = new ArrayList<>();
 
-    if (source.isPdf()) {
+    if (source.isPDF()) {
       this.saveFormat = "jpg";
-      var pdfPageImages = pdfOperation.pdfToImages(source);
-      for (PdfPageImage pdfPageImage : pdfPageImages) {
+      var pdfPageImages = pdfToImages(source.getFile(), this.filename);
+      for (PDFPageImage pdfPageImage : pdfPageImages) {
         this.pageImages.add(pdfPageImage.getImage());
       }
     } else {
@@ -41,14 +43,34 @@ public ImageExtractor(LocalInputSource source, PDFOperation pdfOperation) throws
     }
   }
 
-  /**
-   * Init from a {@link LocalInputSource}.
-   *
-   * @param source The local source.
-   * @throws IOException Throws if the file can't be accessed.
-   */
-  public ImageExtractor(LocalInputSource source) throws IOException {
-    this(source, new PDFBoxApi());
+  public List<PDFPageImage> pdfToImages(byte[] fileBytes, String filename) throws IOException {
+    PDDocument document = Loader.loadPDF(fileBytes);
+    var pdfRenderer = new PDFRenderer(document);
+    List<PDFPageImage> pdfPageImages = new ArrayList<>();
+    for (int i = 0; i < document.getNumberOfPages(); i++) {
+      var imageBuffer = pdfPageToImageBuffer(i, document, pdfRenderer);
+      pdfPageImages.add(new PDFPageImage(imageBuffer, i, filename, "jpg"));
+    }
+    document.close();
+    return pdfPageImages;
+  }
+
+  private BufferedImage pdfPageToImageBuffer(
+      int index,
+      PDDocument document,
+      PDFRenderer pdfRenderer
+  ) throws IOException {
+    PDRectangle bbox = document.getPage(index).getBBox();
+    float dimension = bbox.getWidth() * bbox.getHeight();
+    int dpi;
+    if (dimension < 200000) {
+      dpi = 300;
+    } else if (dimension < 300000) {
+      dpi = 250;
+    } else {
+      dpi = 200;
+    }
+    return pdfRenderer.renderImageWithDPI(index, dpi, ImageType.RGB);
   }
 
   /**

src/main/java/com/mindee/pdf/PdfPageImage.java → src/main/java/com/mindee/image/PDFPageImage.java

@@ -1,4 +1,4 @@
-package com.mindee.pdf;
+package com.mindee.image;
 
 import com.mindee.MindeeException;
 import com.mindee.input.InputSourceUtils;
@@ -16,13 +16,13 @@
  * A page in a PDF extracted as an image.
  */
 @Getter
-public class PdfPageImage {
+public class PDFPageImage {
   private final BufferedImage image;
   private final int originalIndex;
   private final String saveFormat;
   private final String originalFilename;
 
-  public PdfPageImage(
+  public PDFPageImage(
       BufferedImage image,
       int originalIndex,
       String originalFilename,

src/main/java/com/mindee/input/InputSourceUtils.java

@@ -1,12 +1,6 @@
 package com.mindee.input;
 
 import com.mindee.MindeeException;
-import java.io.ByteArrayInputStream;
-import java.io.IOException;
-import org.apache.pdfbox.Loader;
-import org.apache.pdfbox.io.RandomAccessReadBuffer;
-import org.apache.pdfbox.pdmodel.PDDocument;
-import org.apache.pdfbox.text.PDFTextStripper;
 
 /**
  * Utilities for working with files.
@@ -65,46 +59,4 @@ public static String[] splitNameStrict(String filename) throws MindeeException {
     }
     return new String[] { name, extension };
   }
-
-  /**
-   * Returns true if the file is a PDF.
-   */
-  public static boolean isPdf(byte[] fileBytes) {
-    try {
-      Loader.loadPDF(new RandomAccessReadBuffer(new ByteArrayInputStream(fileBytes)));
-    } catch (IOException e) {
-      return false;
-    }
-    return true;
-  }
-
-  /**
-   * Returns true if the source PDF has source text inside. Returns false for images.
-   *
-   * @param fileBytes A byte array representing a PDF.
-   * @return True if at least one character exists in one page.
-   * @throws MindeeException if the file could not be read.
-   */
-  public static boolean hasSourceText(byte[] fileBytes) {
-    try {
-      PDDocument document = Loader
-        .loadPDF(new RandomAccessReadBuffer(new ByteArrayInputStream(fileBytes)));
-      PDFTextStripper stripper = new PDFTextStripper();
-
-      for (int i = 0; i < document.getNumberOfPages(); i++) {
-        stripper.setStartPage(i + 1);
-        stripper.setEndPage(i + 1);
-        String pageText = stripper.getText(document);
-        if (!pageText.trim().isEmpty()) {
-          document.close();
-          return true;
-        }
-      }
-      document.close();
-    } catch (IOException e) {
-      return false;
-    }
-
-    return false;
-  }
 }

src/main/java/com/mindee/input/LocalInputSource.java

@@ -1,9 +1,10 @@
 package com.mindee.input;
 
 import com.mindee.image.ImageCompressor;
-import com.mindee.pdf.PDFBoxApi;
+import com.mindee.pdf.PDFCompression;
 import com.mindee.pdf.PDFCompressor;
-import com.mindee.pdf.PDFOperation;
+import com.mindee.pdf.PDFInputOperation;
+import com.mindee.pdf.PDFInputOperator;
 import java.io.File;
 import java.io.IOException;
 import java.io.InputStream;
@@ -17,14 +18,18 @@
 /**
  * A source document for Mindee API operations.
  */
-public final class LocalInputSource {
+public class LocalInputSource {
 
   @Getter
   private byte[] file;
   @Getter
   private final String filename;
   @Setter
-  private PDFOperation pdfOperation;
+  private PDFInputOperation pdfInputOperator;
+  @Setter
+  private PDFCompression pdfCompressor;
+  // Store here to avoid recalculating every time.
+  private Boolean isPDF;
 
   public LocalInputSource(InputStream file, String filename) throws IOException {
     this.file = IOUtils.toByteArray(file);
@@ -57,11 +62,18 @@ public LocalInputSource(String fileAsBase64, String filename) {
     this.filename = filename;
   }
 
-  public PDFOperation getPdfOperation() {
-    if (this.pdfOperation == null) {
-      this.pdfOperation = new PDFBoxApi();
+  private PDFInputOperation getPDFInputOperator() {
+    if (this.pdfInputOperator == null) {
+      this.pdfInputOperator = new PDFInputOperator();
+    }
+    return this.pdfInputOperator;
+  }
+
+  private PDFCompression getPDFCompressor() {
+    if (this.pdfCompressor == null) {
+      this.pdfCompressor = new PDFCompressor();
     }
-    return this.pdfOperation;
+    return this.pdfCompressor;
   }
 
   /**
@@ -71,10 +83,10 @@ public PDFOperation getPdfOperation() {
    * @throws IOException If an I/O error occurs during the PDF operation.
    */
   public int getPageCount() throws IOException {
-    if (!this.isPdf()) {
+    if (!this.isPDF()) {
       return 1;
     }
-    return getPdfOperation().getNumberOfPages(this.file);
+    return getPDFInputOperator().getPageCount(this.file);
   }
 
   /**
@@ -84,51 +96,58 @@ public int getPageCount() throws IOException {
    * @throws IOException If an I/O error occurs during the PDF operation.
    */
   public void applyPageOptions(PageOptions pageOptions) throws IOException {
-    if (pageOptions != null && this.isPdf()) {
-      this.file = getPdfOperation().split(this.file, pageOptions).getFile();
+    if (pageOptions != null && this.isPDF()) {
+      this.file = getPDFInputOperator().split(this.file, pageOptions).getFile();
     }
   }
 
-  public boolean isPdf() {
-    return InputSourceUtils.isPdf(this.file);
-  }
-
-  public boolean hasSourceText() {
-    return InputSourceUtils.hasSourceText(this.file);
+  /**
+   * Returns true if the file is a PDF.
+   */
+  public boolean isPDF() {
+    if (this.isPDF == null) {
+      this.isPDF = getPDFInputOperator().isPDF(this.file);
+    }
+    return this.isPDF;
   }
 
   public void compress(
-      Integer quality,
+      int quality,
       Integer maxWidth,
       Integer maxHeight,
       Boolean forceSourceText,
       Boolean disableSourceText
   ) throws IOException {
-    if (isPdf()) {
-      this.file = PDFCompressor.compressPdf(this.file, quality, forceSourceText, disableSourceText);
+    if (isPDF()) {
+      this.file = getPDFCompressor()
+        .compressPDF(this.file, quality, forceSourceText, disableSourceText);
     } else {
       this.file = ImageCompressor.compressImage(this.file, quality, maxWidth, maxHeight);
     }
   }
 
   public void compress(
-      Integer quality,
+      int quality,
       Integer maxWidth,
       Integer maxHeight,
       Boolean forceSourceText
   ) throws IOException {
     this.compress(quality, maxWidth, maxHeight, forceSourceText, true);
   }
 
-  public void compress(Integer quality, Integer maxWidth, Integer maxHeight) throws IOException {
-    this.compress(quality, maxWidth, maxHeight, false, true);
+  public void compress(
+      int quality,
+      boolean forceSourceText,
+      boolean disableSourceText
+  ) throws IOException {
+    this.compress(quality, null, null, forceSourceText, disableSourceText);
   }
 
-  public void compress(Integer quality, Integer maxWidth) throws IOException {
-    this.compress(quality, maxWidth, null, false, true);
+  public void compress(int quality, Integer maxWidth, Integer maxHeight) throws IOException {
+    this.compress(quality, maxWidth, maxHeight, false, true);
   }
 
-  public void compress(Integer quality) throws IOException {
+  public void compress(int quality) throws IOException {
     this.compress(quality, null, null, false, true);
   }
 

src/main/java/com/mindee/pdf/BasePDFExtractor.java

@@ -6,7 +6,6 @@
 import java.awt.image.BufferedImage;
 import java.io.ByteArrayInputStream;
 import java.io.ByteArrayOutputStream;
-import java.io.File;
 import java.io.IOException;
 import java.util.ArrayList;
 import java.util.List;
@@ -33,9 +32,9 @@ public class BasePDFExtractor {
    * @param source The local source.
    * @throws IOException Throws if the file can't be accessed.
    */
-  protected BasePDFExtractor(LocalInputSource source) throws IOException {
+  public BasePDFExtractor(LocalInputSource source) throws IOException {
     this.filename = source.getFilename();
-    if (source.isPdf()) {
+    if (source.isPDF()) {
       this.sourcePdf = Loader.loadPDF(source.getFile());
     } else {
       var document = new PDDocument();
@@ -57,15 +56,6 @@ protected BasePDFExtractor(LocalInputSource source) throws IOException {
     }
   }
 
-  /**
-   * Get the number of pages in the PDF file.
-   *
-   * @return The number of pages in the PDF file.
-   */
-  public int getPageCount() {
-    return sourcePdf.getNumberOfPages();
-  }
-
   /**
    * Converts an array to a buffered image.
    *
@@ -106,10 +96,7 @@ public List<ExtractedPDF> extractSubDocuments(
         + splitName[1];
       extractedPDFs
         .add(
-          new ExtractedPDF(
-            Loader.loadPDF(mergePdfPages(this.sourcePdf, pageIndexElement, false)),
-            fieldFilename
-          )
+          new ExtractedPDF(mergePdfPages(this.sourcePdf, pageIndexElement, false), fieldFilename)
         );
     }
     return extractedPDFs;
@@ -149,25 +136,7 @@ private static byte[] createPdfFromExistingPdf(
     return output;
   }
 
-  /**
-   * Merge specified PDF pages together.
-   *
-   * @param file The PDF file.
-   * @param pageNumbers Lit of page numbers to merge together.
-   */
-  public static byte[] mergePdfPages(File file, List<Integer> pageNumbers) throws IOException {
-    PDDocument document = Loader.loadPDF(file);
-    return createPdfFromExistingPdf(document, pageNumbers, true);
-  }
-
-  public static byte[] mergePdfPages(
-      PDDocument document,
-      List<Integer> pageNumbers
-  ) throws IOException {
-    return mergePdfPages(document, pageNumbers, true);
-  }
-
-  public static byte[] mergePdfPages(
+  public byte[] mergePdfPages(
       PDDocument document,
       List<Integer> pageNumbers,
       boolean closeOriginal