OCR integration

.gitignore

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

build-logic/src/main/kotlin/org.jabref.gradle.base.dependency-rules.gradle.kts

@@ -436,6 +436,18 @@ extraJavaModuleInfo {
         requires("java.naming")
         requires("java.sql")
     }
+    module("net.sourceforge.tess4j:tess4j", "net.sourceforge.tess4j") {
+        exportAllPackages()
+        requireAllDefinedDependencies()
+        requires("java.desktop")
+    }
+    module("net.sourceforge.lept4j:lept4j", "net.sourceforge.lept4j") {
+        exportAllPackages()
+        requireAllDefinedDependencies()
+        requires("java.desktop")
+        requires("java.logging")
+    }
+
     module("org.apache.pdfbox:pdfbox", "org.apache.pdfbox") {
         exportAllPackages()
         requireAllDefinedDependencies()
@@ -446,6 +458,32 @@ extraJavaModuleInfo {
         requireAllDefinedDependencies()
         requires("java.xml")
     }
+
+    module("org.apache.pdfbox:pdfbox-tools", "org.apache.pdfbox.tools") {
+        requireAllDefinedDependencies()
+        exportAllPackages()
+        requires("java.desktop")
+    }
+    module("org.apache.pdfbox:pdfbox-debugger", "org.apache.pdfbox.debugger") {
+        requireAllDefinedDependencies()
+        exportAllPackages()
+    }
+    module("org.apache.pdfbox:jbig2-imageio",  "org.apache.pdfbox.jbig2") {
+        exportAllPackages()
+        requireAllDefinedDependencies()
+        requires("java.desktop")
+    }
+    module("org.jboss:jboss-vfs", "org.jboss.vfs") {
+        preserveExisting()
+    }
+    module("org.jboss.logging:jboss-logging", "org.jboss.logging") {
+        preserveExisting()
+    }
+    module("com.github.jai-imageio:jai-imageio-core", "com.github.jaiimageio.core") {
+        exportAllPackages()
+        requireAllDefinedDependencies()
+        requires("java.desktop")
+    }
     module("com.squareup.okio:okio-jvm", "okio") {
         exportAllPackages()
         requireAllDefinedDependencies()

gradle/wrapper/gradle-wrapper.properties

@@ -1,7 +1,6 @@
 distributionBase=GRADLE_USER_HOME
 distributionPath=wrapper/dists
-distributionSha256Sum=bd71102213493060956ec229d946beee57158dbd89d0e62b91bca0fa2c5f3531
-distributionUrl=https\://services.gradle.org/distributions/gradle-8.14.3-bin.zip
+distributionUrl=https\://files.jabref.org/gradle-9.1.0-jabref-14.zip
 networkTimeout=10000
 validateDistributionUrl=true
 zipStoreBase=GRADLE_USER_HOME

jabgui/build.gradle.kts

@@ -80,7 +80,7 @@ dependencies {
 
     implementation ("org.apache.pdfbox:pdfbox")
 
-    // implementation("net.java.dev.jna:jna")
+    implementation("net.java.dev.jna:jna-jpms")
     implementation("net.java.dev.jna:jna-platform")
 
     implementation("org.eclipse.jgit:org.eclipse.jgit")

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

@@ -51,6 +51,7 @@
 import org.jabref.logic.integrity.FieldCheckers;
 import org.jabref.logic.journals.JournalAbbreviationRepository;
 import org.jabref.logic.l10n.Localization;
+import org.jabref.logic.ocr.OcrService;
 import org.jabref.logic.util.TaskExecutor;
 import org.jabref.model.database.BibDatabaseContext;
 import org.jabref.model.entry.BibEntry;
@@ -80,6 +81,8 @@ public class LinkedFilesEditor extends HBox implements FieldEditorFX {
     @Inject private JournalAbbreviationRepository abbreviationRepository;
     @Inject private TaskExecutor taskExecutor;
     @Inject private UndoManager undoManager;
+    @Inject private OcrService ocrService;
+
 
     private LinkedFilesEditorViewModel viewModel;
 
@@ -325,7 +328,9 @@ private void handleItemMouseClick(LinkedFileViewModel linkedFile, MouseEvent eve
                     bibEntry,
                     viewModel,
                     contextCommandFactory,
-                    multiContextCommandFactory
+                    multiContextCommandFactory,
+                    taskExecutor,
+                    ocrService
             );
 
             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 +11,11 @@
 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.ocr.OcrService;
+import org.jabref.logic.util.TaskExecutor;
 import org.jabref.model.database.BibDatabaseContext;
 import org.jabref.model.entry.BibEntry;
 
@@ -25,21 +30,27 @@ public class ContextMenuFactory {
     private final LinkedFilesEditorViewModel viewModel;
     private final SingleContextCommandFactory singleCommandFactory;
     private final MultiContextCommandFactory multiCommandFactory;
+    private final TaskExecutor taskExecutor;
+    private final OcrService ocrService;
 
     public ContextMenuFactory(DialogService dialogService,
                               GuiPreferences preferences,
                               BibDatabaseContext databaseContext,
                               ObservableOptionalValue<BibEntry> bibEntry,
                               LinkedFilesEditorViewModel viewModel,
                               SingleContextCommandFactory singleCommandFactory,
-                              MultiContextCommandFactory multiCommandFactory) {
+                              MultiContextCommandFactory multiCommandFactory,
+                              TaskExecutor taskExecutor,
+                              OcrService ocrService) {
         this.dialogService = dialogService;
         this.preferences = preferences;
         this.databaseContext = databaseContext;
         this.bibEntry = bibEntry;
         this.viewModel = viewModel;
         this.singleCommandFactory = singleCommandFactory;
         this.multiCommandFactory = multiCommandFactory;
+        this.taskExecutor = taskExecutor;
+        this.ocrService = ocrService;
     }
 
     public ContextMenu createForSelection(ObservableList<LinkedFileViewModel> selectedFiles) {
@@ -86,9 +97,46 @@ 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,
+                ocrService
+        );
+
+        // 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 +1,115 @@
+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.OcrResult;
+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;
+
+import org.slf4j.Logger;
+import org.slf4j.LoggerFactory;
+
+/**
+ * Action for performing OCR (Optical Character Recognition) on linked PDF files.
+ */
+public class OcrAction extends SimpleCommand {
+    private static final Logger LOGGER = LoggerFactory.getLogger(OcrAction.class);
+
+    private final LinkedFile linkedFile;
+    private final BibDatabaseContext databaseContext;
+    private final DialogService dialogService;
+    private final FilePreferences filePreferences;
+    private final TaskExecutor taskExecutor;
+    private final OcrService ocrService;
+
+    public OcrAction(LinkedFile linkedFile,
+                     BibDatabaseContext databaseContext,
+                     DialogService dialogService,
+                     FilePreferences filePreferences,
+                     TaskExecutor taskExecutor,
+                     OcrService ocrService) {
+        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()
+        );
+        this.ocrService = ocrService;
+    }
+
+    @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<OcrResult> task = BackgroundTask.wrap(() -> {
+                                                           return ocrService.performOcr(filePath.get());
+                                                       })
+                                                       .showToUser(true)  // Show in task list
+                                                       .withInitialMessage(Localization.lang("Performing OCR on %0", linkedFile.getLink()));
+
+        task.onSuccess(result -> {
+                // Use pattern matching with the sealed class
+                switch (result) {
+                    case OcrResult.Success success -> {
+                        String extractedText = success.text();
+                        if (extractedText.isEmpty()) {
+                            dialogService.showInformationDialogAndWait(
+                                    Localization.lang("OCR Complete"),
+                                    Localization.lang("No text was found in the PDF.")
+                            );
+                        } else {
+                            // Show preview
+                            String preview = extractedText.length() > 1000
+                                    ? extractedText.substring(0, 1000) + "..."
+                                    : extractedText;
+
+                            dialogService.showInformationDialogAndWait(
+                                    Localization.lang("OCR Result"),
+                                    preview
+                            );
+                        }
+                    }
+                    case OcrResult.Failure failure -> {
+                        dialogService.showErrorDialogAndWait(
+                                Localization.lang("OCR failed"),
+                                failure.errorMessage()
+                        );
+                    }
+                }
+            })
+            .onFailure(exception -> {
+                dialogService.showErrorDialogAndWait(
+                        Localization.lang("OCR failed"),
+                        exception.getMessage()
+                );
+            })
+            .executeWith(taskExecutor);
+    }
+}