OCR integration

.gitignore

@@ -581,3 +581,5 @@ CHANGELOG.html
 
 # some strange gradle/IntelliJ extension
 extension 'reporting' property 'baseDirectory'
+
+tessdata/

jabgui/src/main/java/org/jabref/gui/fieldeditors/LinkedFilesEditor.java

@@ -325,7 +325,8 @@ private void handleItemMouseClick(LinkedFileViewModel linkedFile, MouseEvent eve
                     bibEntry,
                     viewModel,
                     contextCommandFactory,
-                    multiContextCommandFactory
+                    multiContextCommandFactory,
+                    taskExecutor
             );
 
             ContextMenu contextMenu = contextMenuFactory.createForSelection(listView.getSelectionModel().getSelectedItems());

jabgui/src/main/java/org/jabref/gui/fieldeditors/contextmenu/ContextMenuFactory.java

@@ -2,6 +2,7 @@
 
 import javafx.collections.ObservableList;
 import javafx.scene.control.ContextMenu;
+import javafx.scene.control.MenuItem;
 import javafx.scene.control.SeparatorMenuItem;
 
 import org.jabref.gui.DialogService;
@@ -10,7 +11,10 @@
 import org.jabref.gui.copyfiles.CopySingleFileAction;
 import org.jabref.gui.fieldeditors.LinkedFileViewModel;
 import org.jabref.gui.fieldeditors.LinkedFilesEditorViewModel;
+import org.jabref.gui.linkedfile.OcrAction;
 import org.jabref.gui.preferences.GuiPreferences;
+import org.jabref.logic.l10n.Localization;
+import org.jabref.logic.util.TaskExecutor;
 import org.jabref.model.database.BibDatabaseContext;
 import org.jabref.model.entry.BibEntry;
 
@@ -25,21 +29,24 @@ public class ContextMenuFactory {
     private final LinkedFilesEditorViewModel viewModel;
     private final SingleContextCommandFactory singleCommandFactory;
     private final MultiContextCommandFactory multiCommandFactory;
+    private final TaskExecutor taskExecutor;
 
     public ContextMenuFactory(DialogService dialogService,
                               GuiPreferences preferences,
                               BibDatabaseContext databaseContext,
                               ObservableOptionalValue<BibEntry> bibEntry,
                               LinkedFilesEditorViewModel viewModel,
                               SingleContextCommandFactory singleCommandFactory,
-                              MultiContextCommandFactory multiCommandFactory) {
+                              MultiContextCommandFactory multiCommandFactory,
+                              TaskExecutor taskExecutor) {
         this.dialogService = dialogService;
         this.preferences = preferences;
         this.databaseContext = databaseContext;
         this.bibEntry = bibEntry;
         this.viewModel = viewModel;
         this.singleCommandFactory = singleCommandFactory;
         this.multiCommandFactory = multiCommandFactory;
+        this.taskExecutor = taskExecutor;
     }
 
     public ContextMenu createForSelection(ObservableList<LinkedFileViewModel> selectedFiles) {
@@ -86,9 +93,45 @@ private ContextMenu createContextMenuForFile(LinkedFileViewModel linkedFile) {
                 factory.createMenuItem(StandardActions.DELETE_FILE, singleCommandFactory.build(StandardActions.DELETE_FILE, linkedFile))
         );
 
+        // Add OCR menu item for PDF files
+        if (linkedFile.getFile().getFileType().equalsIgnoreCase("pdf")) {
+            menu.getItems().add(new SeparatorMenuItem());
+
+            MenuItem ocrItem = createOcrMenuItem(linkedFile);
+            menu.getItems().add(ocrItem);
+        }
+
         return menu;
     }
 
+    /**
+     * Creates the OCR menu item for a PDF file.
+     * The menu item is only enabled if the PDF file exists on disk.
+     *
+     * @param linkedFile The linked PDF file
+     * @return MenuItem configured for OCR action
+     */
+    private MenuItem createOcrMenuItem(LinkedFileViewModel linkedFile) {
+        MenuItem ocrItem = new MenuItem(Localization.lang("Extract text (OCR)"));
+
+        // Create the OCR action
+        OcrAction ocrAction = new OcrAction(
+                linkedFile.getFile(),
+                databaseContext,
+                dialogService,
+                preferences.getFilePreferences(),
+                taskExecutor
+        );
+
+        // Set the action to execute when clicked
+        ocrItem.setOnAction(event -> ocrAction.execute());
+
+        // Disable if the action is not executable (file doesn't exist)
+        ocrItem.disableProperty().bind(ocrAction.executableProperty().not());
+
+        return ocrItem;
+    }
+
     @FunctionalInterface
     public interface SingleContextCommandFactory {
         ContextAction build(StandardActions action, LinkedFileViewModel file);

jabgui/src/main/java/org/jabref/gui/linkedfile/OcrAction.java

@@ -0,0 +1,106 @@
+package org.jabref.gui.linkedfile;
+
+import org.jabref.gui.DialogService;
+import org.jabref.gui.StateManager;
+import org.jabref.gui.actions.Action;
+import org.jabref.gui.actions.ActionHelper;
+import org.jabref.gui.actions.SimpleCommand;
+import org.jabref.logic.util.BackgroundTask;
+import org.jabref.logic.util.TaskExecutor;
+import org.jabref.logic.l10n.Localization;
+import org.jabref.logic.ocr.OcrService;
+import org.jabref.logic.ocr.OcrException;
+import org.jabref.model.database.BibDatabaseContext;
+import org.jabref.model.entry.LinkedFile;
+import org.jabref.logic.FilePreferences;
+
+import java.nio.file.Path;
+import java.util.Optional;
+
+/**
+ * Action for performing OCR (Optical Character Recognition) on linked PDF files.
+ * <p>
+ * This action extracts text content from PDF files that are attached to BibTeX entries.
+ * It runs the OCR process in a background thread to keep the UI responsive and provides
+ * user feedback through dialogs and notifications.
+ * <p>
+ * The action follows JabRef's command pattern and can be triggered from context menus.
+ * It includes built-in validation to ensure it's only enabled for PDF files that exist on disk.
+ *
+ * @see OcrService
+ * @see org.jabref.gui.actions.SimpleCommand
+ */
+
+// Personal Note: Add more doc in between later
+
+public class OcrAction extends SimpleCommand {
+
+    private final LinkedFile linkedFile;
+    private final BibDatabaseContext databaseContext;
+    private final DialogService dialogService;
+    private final FilePreferences filePreferences;
+    private final TaskExecutor taskExecutor;
+
+    public OcrAction(LinkedFile linkedFile,
+                     BibDatabaseContext databaseContext,
+                     DialogService dialogService,
+                     FilePreferences filePreferences,
+                     TaskExecutor taskExecutor) {
+        this.linkedFile = linkedFile;
+        this.databaseContext = databaseContext;
+        this.dialogService = dialogService;
+        this.filePreferences = filePreferences;
+        this.taskExecutor = taskExecutor;
+
+        // Only executable for existing PDF files
+        this.executable.set(
+                linkedFile.getFileType().equalsIgnoreCase("pdf") &&
+                        linkedFile.findIn(databaseContext, filePreferences).isPresent()
+        );
+    }
+
+    @Override
+    public void execute() {
+        Optional<Path> filePath = linkedFile.findIn(databaseContext, filePreferences);
+
+        if (filePath.isEmpty()) {
+            dialogService.showErrorDialogAndWait(
+                    Localization.lang("File not found"),
+                    Localization.lang("Could not locate the PDF file on disk.")
+            );
+            return;
+        }
+
+        dialogService.notify(Localization.lang("Performing OCR..."));
+
+        BackgroundTask.wrap(() -> {
+                          OcrService ocrService = new OcrService();
+                          return ocrService.performOcr(filePath.get());
+                      })
+                      .onSuccess(extractedText -> {
+                          if (extractedText.isEmpty()) {
+                              dialogService.showInformationDialogAndWait(
+                                      Localization.lang("OCR Complete"),
+                                      Localization.lang("No text was found in the PDF.")
+                              );
+                          } else {
+                              // For now, just show preview
+                              String preview = extractedText.length() > 1000
+                                      ? extractedText.substring(0, 1000) + "..."
+                                      : extractedText;
+
+                              dialogService.showInformationDialogAndWait(
+                                      Localization.lang("OCR Result"),
+                                      preview
+                              );
+                          }
+                      })
+                      .onFailure(exception -> {
+                          dialogService.showErrorDialogAndWait(
+                                  Localization.lang("OCR failed"),
+                                  exception.getMessage()
+                          );
+                      })
+                      .executeWith(taskExecutor);
+    }
+}

jabgui/src/test/java/org/jabref/gui/fieldeditors/contextmenu/ContextMenuFactoryTest.java

@@ -11,6 +11,7 @@
 import org.jabref.gui.fieldeditors.LinkedFileViewModel;
 import org.jabref.gui.fieldeditors.LinkedFilesEditorViewModel;
 import org.jabref.gui.preferences.GuiPreferences;
+import org.jabref.logic.util.TaskExecutor;
 import org.jabref.model.database.BibDatabaseContext;
 import org.jabref.model.entry.BibEntry;
 import org.jabref.model.entry.LinkedFile;
@@ -41,6 +42,7 @@ public class ContextMenuFactoryTest {
     private ContextMenuFactory factory;
     private ContextMenuFactory.SingleContextCommandFactory singleCommandFactory;
     private ContextMenuFactory.MultiContextCommandFactory multiCommandFactory;
+    private TaskExecutor taskExecutor;
 
     @BeforeAll
     public static void initToolkit() {
@@ -78,7 +80,8 @@ public void setUp() {
                 bibEntry,
                 viewModel,
                 singleCommandFactory,
-                multiCommandFactory
+                multiCommandFactory,
+                taskExecutor
         );
     }
 

jablib/build.gradle.kts

@@ -253,6 +253,9 @@ dependencies {
     // Required for LocalizationConsistencyTest
     testImplementation("org.testfx:testfx-core:4.0.16-alpha")
     testImplementation("org.testfx:testfx-junit5:4.0.16-alpha")
+
+    // OCR support
+    implementation("net.sourceforge.tess4j:tess4j:5.15.0")
 }
 /*
 jacoco {

jablib/src/main/java/module-info.java

@@ -105,6 +105,7 @@
     exports org.jabref.logic.crawler;
     exports org.jabref.logic.git;
     exports org.jabref.logic.pseudonymization;
+    exports org.jabref.logic.ocr;
     exports org.jabref.logic.citation.repository;
 
     requires java.base;
@@ -252,5 +253,6 @@
     requires mslinks;
     requires org.antlr.antlr4.runtime;
     requires org.libreoffice.uno;
+    requires tess4j;
     // endregion
 }

jablib/src/main/java/org/jabref/logic/ocr/OcrException.java

@@ -0,0 +1,28 @@
+package org.jabref.logic.ocr;
+
+/**
+ * Exception thrown when OCR operations fail.
+ * This exception wraps lower-level OCR engine exceptions to provide
+ * a consistent interface for error handling throughout JabRef.
+ */
+public class OcrException extends Exception {
+
+    /**
+     * Constructs an OcrException with a message and underlying cause.
+     *
+     * @param message Descriptive error message
+     * @param cause The underlying exception that caused this error
+     */
+    public OcrException(String message, Throwable cause) {
+        super(message, cause);
+    }
+
+    /**
+     * Constructs an OcrException with only a message.
+     *
+     * @param message Descriptive error message
+     */
+    public OcrException(String message) {
+        super(message);
+    }
+}

jablib/src/main/java/org/jabref/logic/ocr/OcrService.java

@@ -0,0 +1,78 @@
+package org.jabref.logic.ocr;
+
+import net.sourceforge.tess4j.Tesseract;
+import net.sourceforge.tess4j.TesseractException;
+import org.jabref.model.strings.StringUtil;  // JabRef utility class
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+import java.io.File;
+import java.nio.file.Path;
+
+/**
+ * Service for performing Optical Character Recognition (OCR) on PDF files.
+ * This class provides a high-level interface to OCR functionality,
+ * abstracting away the specific OCR engine implementation details.
+ */
+public class OcrService {
+    private static final Logger LOGGER = LoggerFactory.getLogger(OcrService.class);
+
+    // The OCR engine instance
+    private final Tesseract tesseract;
+
+    /**
+     * Constructs a new OcrService with default settings.
+     * Currently uses Tesseract with English language support.
+     */
+    public OcrService() {
+        this.tesseract = new Tesseract();
+
+        // Configure Tesseract
+        tesseract.setLanguage("eng");
+
+        // TODO: This path needs to be configurable and bundled properly
+        // For now, we'll use a relative path that works during development
+        tesseract.setDatapath("tessdata");
+
+        LOGGER.debug("Initialized OcrService with Tesseract");
+    }
+
+    /**
+     * Performs OCR on a PDF file and returns the extracted text.
+     *
+     * @param pdfPath Path to the PDF file to process
+     * @return The extracted text, or empty string if no text found
+     * @throws OcrException if OCR processing fails
+     */
+    public String performOcr(Path pdfPath) throws OcrException {
+        // Validate input
+        if (pdfPath == null) {
+            throw new OcrException("PDF path cannot be null");
+        }
+
+        File pdfFile = pdfPath.toFile();
+        if (!pdfFile.exists()) {
+            throw new OcrException("PDF file does not exist: " + pdfPath);
+        }
+
+        try {
+            LOGGER.info("Starting OCR for file: {}", pdfFile.getName());
+
+            // Perform OCR
+            String result = tesseract.doOCR(pdfFile);
+
+            // Clean up the result (remove extra whitespace, etc.)
+            result = StringUtil.isBlank(result) ? "" : result.trim();
+
+            LOGGER.info("OCR completed successfully. Extracted {} characters", result.length());
+            return result;
+
+        } catch (TesseractException e) {
+            LOGGER.error("OCR failed for file: {}", pdfFile.getName(), e);
+            throw new OcrException(
+                    "Failed to perform OCR on file: " + pdfFile.getName() +
+                            ". Error: " + e.getMessage(), e
+            );
+        }
+    }
+}

jablib/src/main/resources/l10n/JabRef_en.properties

@@ -316,6 +316,14 @@ Extract\ references\ from\ file\ (online)=Extract references from file (online)
 Extract\ References\ (offline)=Extract References (offline)
 Extract\ References\ (online)=Extract References (online)
 
+Extract\ text\ (OCR)=Extract text (OCR)
+Performing\ OCR...=Performing OCR...
+OCR\ Complete=OCR Complete
+OCR\ Result=OCR Result
+OCR\ failed=OCR failed
+No\ text\ was\ found\ in\ the\ PDF.=No text was found in the PDF.
+Could\ not\ locate\ the\ PDF\ file\ on\ disk.=Could not locate the PDF file on disk.
+
 Processing...=Processing...
 Processing\ "%0"...=Processing "%0"...
 Processing\ PDF(s)=Processing PDF(s)