Get rid of deprecated OcrPdfCreator#getLogicalTree

DEVSIX-8570

Author: AnhelinaM

pdfocr-api/src/main/java/com/itextpdf/pdfocr/IOcrEngine.java

@@ -85,4 +85,11 @@ public interface IOcrEngine {
      * @param ocrProcessContext ocr processing context
      */
     void createTxtFile(List<File> inputImages, File txtFile, OcrProcessContext ocrProcessContext);
+
+    /**
+     * Checks whether tagging is supported by the OCR engine.
+     *
+     * @return {@code true} if tagging is supported by the engine, {@code false} otherwise
+     */
+    boolean isTaggingSupported();
 }

pdfocr-api/src/main/java/com/itextpdf/pdfocr/OcrPdfCreator.java

@@ -129,6 +129,9 @@ public OcrPdfCreator(final IOcrEngine ocrEngine) {
      */
     public OcrPdfCreator(final IOcrEngine ocrEngine,
             final OcrPdfCreatorProperties ocrPdfCreatorProperties) {
+        if (ocrPdfCreatorProperties.isTagged() && !ocrEngine.isTaggingSupported()) {
+            throw new PdfOcrException(PdfOcrExceptionMessageConstant.TAGGING_IS_NOT_SUPPORTED);
+        }
         setOcrEngine(ocrEngine);
         setOcrPdfCreatorProperties(ocrPdfCreatorProperties);
     }
@@ -471,11 +474,7 @@ private void addToCanvas(final PdfDocument pdfDocument,
                 // Logical tree, a list of top items, children can be retrieved out of them
                 List<LogicalStructureTreeItem> logicalTree = new ArrayList<>();
                 // A map of leaf LogicalStructureTreeItem's to TextInfo's attached to these leaves
-                Map<LogicalStructureTreeItem, List<TextInfo>> leavesTextInfos = new HashMap<>();
-                final boolean taggedSupported = getLogicalTree(pageText, logicalTree, leavesTextInfos);
-                if (!taggedSupported) {
-                    throw new PdfOcrException(PdfOcrExceptionMessageConstant.TAGGING_IS_NOT_SUPPORTED);
-                }
+                Map<LogicalStructureTreeItem, List<TextInfo>> leavesTextInfos = getLogicalTree(pageText, logicalTree);
                 pdfDocument.setTagged();
 
                 // Create a map of TextInfo to tag pointers meanwhile creating the required tags.
@@ -634,19 +633,12 @@ private void addImageToCanvas(final ImageData imageData,
         }
     }
 
-    /**
-     * @return {@code true} if tagging supported by the engine.
-     * @deprecated In next major version we need to add boolean taggingSupported() method into IOcrEngine
-     * and throw exception in OcrPdfCreator constructor if taggingSupported() returns false but
-     * OcrPdfCreatorProperties.getTagged returns true.
-     */
-    @Deprecated
-    private static boolean getLogicalTree(List<TextInfo> textInfos,
-            List<LogicalStructureTreeItem> logicalStructureTreeItems,
-            Map<LogicalStructureTreeItem, List<TextInfo>> leavesTextInfos) {
-        boolean taggedSupported = false;
+    private static Map<LogicalStructureTreeItem, List<TextInfo>> getLogicalTree(
+            List<TextInfo> textInfos, List<LogicalStructureTreeItem> logicalStructureTreeItems) {
+
+        Map<LogicalStructureTreeItem, List<TextInfo>> leavesTextInfos = new HashMap<>();
         if (textInfos == null) {
-            return taggedSupported;
+            return leavesTextInfos;
         }
 
         for (TextInfo textInfo : textInfos) {
@@ -656,7 +648,6 @@ private static boolean getLogicalTree(List<TextInfo> textInfos,
                 continue;
             } else if (structTreeItem != null) {
                 topParent = getTopParent(structTreeItem);
-                taggedSupported = true;
             } else {
                 structTreeItem = new LogicalStructureTreeItem();
                 textInfo.setLogicalStructureTreeItem(structTreeItem);
@@ -675,7 +666,7 @@ private static boolean getLogicalTree(List<TextInfo> textInfos,
             }
         }
 
-        return taggedSupported;
+        return leavesTextInfos;
     }
 
     private static LogicalStructureTreeItem getTopParent(LogicalStructureTreeItem structInfo) {

pdfocr-api/src/test/java/com/itextpdf/pdfocr/ApiTest.java

@@ -269,17 +269,14 @@ public void testTableStructureTree() throws IOException, InterruptedException {
     }
 
     @Test
-    @LogMessages(messages = @LogMessage(messageTemplate = PdfOcrExceptionMessageConstant.CANNOT_CREATE_PDF_DOCUMENT,
-            logLevel = LogLevelConstants.ERROR))
     public void testTaggingNotSupported() {
         String input = PdfHelper.getImagesTestDirectory() + "numbers_01.jpg";
         String pdfPath = PdfHelper.getTargetDirectory() + "taggingNotSupported.pdf";
 
         Exception e = Assertions.assertThrows(PdfOcrException.class,
                 () -> PdfHelper.createPdf(pdfPath, new File(input), new OcrPdfCreatorProperties().setTagged(true))
         );
-        Assertions.assertEquals(MessageFormatUtil.format(PdfOcrExceptionMessageConstant.CANNOT_CREATE_PDF_DOCUMENT,
-                PdfOcrExceptionMessageConstant.TAGGING_IS_NOT_SUPPORTED), e.getMessage());
+        Assertions.assertEquals(PdfOcrExceptionMessageConstant.TAGGING_IS_NOT_SUPPORTED, e.getMessage());
     }
 
     static class NotImplementedImageRotationHandler implements IImageRotationHandler {

pdfocr-api/src/test/java/com/itextpdf/pdfocr/helpers/CustomOcrEngine.java

@@ -47,8 +47,7 @@ public CustomOcrEngine(OcrEngineProperties ocrEngineProperties) {
 
     @Override
     public Map<Integer, List<TextInfo>> doImageOcr(File input) {
-        Map<Integer, List<TextInfo>> result =
-                new HashMap<Integer, List<TextInfo>>();
+        Map<Integer, List<TextInfo>> result = new HashMap<Integer, List<TextInfo>>();
         String text = PdfHelper.DEFAULT_TEXT;
         if (input.getAbsolutePath().contains(PdfHelper.THAI_IMAGE_NAME)) {
             text = PdfHelper.THAI_TEXT;
@@ -72,6 +71,10 @@ public void createTxtFile(List<File> inputImages, File txtFile, OcrProcessContex
 
     }
 
+    @Override
+    public boolean isTaggingSupported() {
+        return false;
+    }
 
     public OcrEngineProperties getOcrEngineProperties() {
         return ocrEngineProperties;

pdfocr-api/src/test/java/com/itextpdf/pdfocr/helpers/CustomProductAwareOcrEngine.java

@@ -24,7 +24,6 @@ This file is part of the iText (R) project.
 
 import com.itextpdf.commons.actions.contexts.IMetaInfo;
 import com.itextpdf.commons.actions.data.ProductData;
-import com.itextpdf.kernel.geom.Rectangle;
 import com.itextpdf.pdfocr.IOcrEngine;
 import com.itextpdf.pdfocr.IProductAware;
 import com.itextpdf.pdfocr.OcrEngineProperties;
@@ -34,7 +33,6 @@ This file is part of the iText (R) project.
 
 import java.io.File;
 import java.util.Collections;
-import java.util.HashMap;
 import java.util.List;
 import java.util.Map;
 
@@ -64,6 +62,10 @@ public void createTxtFile(List<File> inputImages, File txtFile, OcrProcessContex
 
     }
 
+    @Override
+    public boolean isTaggingSupported() {
+        return true;
+    }
 
     public OcrEngineProperties getOcrEngineProperties() {
         return null;

pdfocr-api/src/test/java/com/itextpdf/pdfocr/helpers/TestStructureDetectionOcrEngine.java

@@ -38,7 +38,6 @@ This file is part of the iText (R) project.
 import java.util.HashMap;
 import java.util.List;
 import java.util.Map;
-import org.junit.jupiter.api.Test;
 
 public class TestStructureDetectionOcrEngine implements IOcrEngine {
 
@@ -101,4 +100,9 @@ public void createTxtFile(List<File> inputImages, File txtFile) {
     @Override
     public void createTxtFile(List<File> inputImages, File txtFile, OcrProcessContext ocrProcessContext) {
     }
+
+    @Override
+    public boolean isTaggingSupported() {
+        return true;
+    }
 }

pdfocr-tesseract4/src/main/java/com/itextpdf/pdfocr/tesseract4/AbstractTesseract4OcrEngine.java

@@ -60,9 +60,10 @@ This file is part of the iText (R) project.
 /**
  * The implementation of {@link IOcrEngine}.
  *
+ * <p>
  * This class provides possibilities to perform OCR, to read data from input
  * files and to return contained text in the required format.
- * Also there are possibilities to use features of "tesseract"
+ * Also, there are possibilities to use features of "tesseract"
  * (optical character recognition engine for various operating systems).
  */
 public abstract class AbstractTesseract4OcrEngine implements IOcrEngine, IProductAware {
@@ -379,6 +380,11 @@ public ProductData getProductData() {
         return PdfOcrTesseract4ProductData.getInstance();
     }
 
+    @Override
+    public boolean isTaggingSupported() {
+        return false;
+    }
+
     /**
      * Performs tesseract OCR using command line tool
      * or a wrapper for Tesseract OCR API.

pdfocr-tesseract4/src/main/java/com/itextpdf/pdfocr/tesseract4/Tesseract4ExecutableOcrEngine.java

@@ -22,7 +22,6 @@ This file is part of the iText (R) project.
  */
 package com.itextpdf.pdfocr.tesseract4;
 
-import com.itextpdf.commons.actions.EventManager;
 import com.itextpdf.commons.actions.confirmations.ConfirmEvent;
 import com.itextpdf.commons.actions.confirmations.EventConfirmationType;
 import com.itextpdf.commons.utils.MessageFormatUtil;
@@ -49,6 +48,7 @@ This file is part of the iText (R) project.
 /**
  * The implementation of {@link AbstractTesseract4OcrEngine} for tesseract OCR.
  *
+ * <p>
  * This class provides possibilities to use features of "tesseract" CL tool
  * (optical character recognition engine for various operating systems).
  * Please note that it's assumed that "tesseract" has already been

pdfocr-tesseract4/src/main/java/com/itextpdf/pdfocr/tesseract4/Tesseract4LibOcrEngine.java

@@ -22,7 +22,6 @@ This file is part of the iText (R) project.
  */
 package com.itextpdf.pdfocr.tesseract4;
 
-import com.itextpdf.commons.actions.EventManager;
 import com.itextpdf.commons.actions.confirmations.ConfirmEvent;
 import com.itextpdf.commons.actions.confirmations.EventConfirmationType;
 import com.itextpdf.commons.utils.MessageFormatUtil;
@@ -53,9 +52,11 @@ This file is part of the iText (R) project.
 /**
  * The implementation of {@link AbstractTesseract4OcrEngine} for tesseract OCR.
  *
+ * <p>
  * This class provides possibilities to use features of "tesseract"
  * using tess4j.
  *
+ * <p>
  * Please note that this class is not thread-safe, in other words this Tesseract engine cannot
  * be used for multithreaded processing. You should create one instance per thread
  */

pdfocr-tesseract4/src/test/java/com/itextpdf/pdfocr/tesseract4/ApiTest.java

@@ -24,7 +24,11 @@ This file is part of the iText (R) project.
 
 import com.itextpdf.commons.utils.MessageFormatUtil;
 import com.itextpdf.pdfocr.IntegrationTestHelper;
+import com.itextpdf.pdfocr.OcrPdfCreator;
+import com.itextpdf.pdfocr.OcrPdfCreatorProperties;
 import com.itextpdf.pdfocr.TextInfo;
+import com.itextpdf.pdfocr.exceptions.PdfOcrException;
+import com.itextpdf.pdfocr.exceptions.PdfOcrExceptionMessageConstant;
 import com.itextpdf.pdfocr.tesseract4.exceptions.PdfOcrTesseract4Exception;
 import com.itextpdf.pdfocr.tesseract4.exceptions.PdfOcrTesseract4ExceptionMessageConstant;
 import com.itextpdf.pdfocr.tesseract4.logs.Tesseract4LogMessageConstant;
@@ -152,4 +156,21 @@ public void testDetectAndFixBrokenBBoxes() throws IOException {
         Assertions.assertEquals(162.75, (float)textInfo.getBboxRect().getTop(), 0.1);
     }
 
+    @Test
+    public void testTaggingNotSupportedForTesseract4ExecutableOcrEngine() {
+        Exception e = Assertions.assertThrows(PdfOcrException.class,
+                () -> new OcrPdfCreator(new Tesseract4ExecutableOcrEngine(new Tesseract4OcrEngineProperties()),
+                        new OcrPdfCreatorProperties().setTagged(true))
+        );
+        Assertions.assertEquals(PdfOcrExceptionMessageConstant.TAGGING_IS_NOT_SUPPORTED, e.getMessage());
+    }
+
+    @Test
+    public void testTaggingNotSupportedForTesseract4LibOcrEngine() {
+        Exception e = Assertions.assertThrows(PdfOcrException.class,
+                () -> new OcrPdfCreator(new Tesseract4LibOcrEngine(new Tesseract4OcrEngineProperties()),
+                        new OcrPdfCreatorProperties().setTagged(true))
+        );
+        Assertions.assertEquals(PdfOcrExceptionMessageConstant.TAGGING_IS_NOT_SUPPORTED, e.getMessage());
+    }
 }