Move language pack locations to AppData

Core/src/org/sleuthkit/autopsy/core/Installer.java

@@ -20,6 +20,7 @@ package org.sleuthkit.autopsy.core;
 
 import java.awt.Cursor;
 import java.io.File;
+import java.io.IOException;
 import java.util.ArrayList;
 import java.util.List;
 import java.util.concurrent.Callable;
@@ -29,6 +30,8 @@ import java.util.logging.Handler;
 import java.util.logging.Level;
 import javafx.application.Platform;
 import javafx.embed.swing.JFXPanel;
+import org.apache.commons.io.FileUtils;
+import org.openide.modules.InstalledFileLocator;
 import org.openide.modules.ModuleInstall;
 import org.openide.util.NbBundle;
 import org.openide.windows.WindowManager;
@@ -286,11 +289,33 @@ public class Installer extends ModuleInstall {
         pythonModulesDir.mkdir();
     }
     
+    /**
+     * Make a folder in the config directory for Ocr Language Packs if one does
+     * not exist.
+     */
+    private static void ensureOcrLanguagePacksFolderExists() {
+        File ocrLanguagePacksDir = new File(PlatformUtil.getOcrLanguagePacksPath());
+        boolean createDirectory = ocrLanguagePacksDir.mkdir();
+        
+        //If the directory did not exist, copy the tessdata folder over so we 
+        //support english.
+        if(createDirectory) {
+            File tessdataDir = InstalledFileLocator.getDefault().locate(
+                    "Tesseract-OCR/tessdata", Installer.class.getPackage().getName(), false);
+            try {
+                FileUtils.copyDirectory(tessdataDir, ocrLanguagePacksDir);
+            } catch (IOException ex) {
+                logger.log(Level.SEVERE, "Copying over default language packs for Tesseract failed.", ex);
+            }
+        }
+    }
+
     @Override
     public void restored() {
         super.restored();
         ensurePythonModulesFolderExists();
         ensureClassifierFolderExists();
+        ensureOcrLanguagePacksFolderExists();
         initJavaFx();
         for (ModuleInstall mi : packageInstallers) {
             try {

Core/src/org/sleuthkit/autopsy/coreutils/PlatformUtil.java

@@ -52,6 +52,7 @@ public class PlatformUtil {
 
     private static final String PYTHON_MODULES_SUBDIRECTORY = "python_modules"; //NON-NLS
     private static final String CLASSIFIERS_SUBDIRECTORY = "object_detection_classifiers"; //NON-NLS
+    private static final String OCR_LANGUAGE_SUBDIRECTORY = "ocr_language_packs"; //NON-NLS
     private static String javaPath = null;
     public static final String OS_NAME_UNKNOWN = NbBundle.getMessage(PlatformUtil.class, "PlatformUtil.nameUnknown");
     public static final String OS_VERSION_UNKNOWN = NbBundle.getMessage(PlatformUtil.class, "PlatformUtil.verUnknown");
@@ -117,6 +118,15 @@ public class PlatformUtil {
         return getUserDirectory().getAbsolutePath() + File.separator + PYTHON_MODULES_SUBDIRECTORY;
     }
     
+    /**
+     * Get root path where the user's Ocr language packs are stored.
+     * 
+     * @return Absolute path to the Ocr language packs root directory.
+     */
+    public static String getOcrLanguagePacksPath() {
+        return getUserDirectory().getAbsolutePath() + File.separator + OCR_LANGUAGE_SUBDIRECTORY;
+    }
+
     /**
      * Get root path where the user's object detection classifiers are stored.
      * 

Core/src/org/sleuthkit/autopsy/textreaders/TikaTextExtractor.java

@@ -202,14 +202,9 @@ final class TikaTextExtractor extends TextExtractor {
                 TesseractOCRConfig ocrConfig = new TesseractOCRConfig();
                 String tesseractFolder = TESSERACT_PATH.getParent();
                 ocrConfig.setTesseractPath(tesseractFolder);
-                /*
-                 * Tesseract expects language data packs to be in a subdirectory
-                 * of tesseractFolder, in a folder called "tessdata". If they
-                 * are stored somewhere else, use
-                 * ocrConfig.setTessdataPath(String tessdataPath) to point to
-                 * them
-                 */
+                
                 ocrConfig.setLanguage(LANGUAGE_PACKS);
+                ocrConfig.setTessdataPath(PlatformUtil.getOcrLanguagePacksPath());
                 parseContext.set(TesseractOCRConfig.class, ocrConfig);
 
                 stream = new ReadContentInputStream(content);
@@ -292,6 +287,7 @@ final class TikaTextExtractor extends TextExtractor {
             process.command(executeablePath,
                     String.format("\"%s\"", inputFile.getAbsolutePath()),
                     String.format("\"%s\"", outputFilePath),
+                    "--tessdata-dir", PlatformUtil.getOcrLanguagePacksPath(),
                     //language pack command flag
                     "-l", LANGUAGE_PACKS);
 
@@ -450,10 +446,7 @@ final class TikaTextExtractor extends TextExtractor {
      * @return String of all language packs available for Tesseract to use
      */
     private static String getLanguagePacks() {
-        File languagePackRootDir = new File(TESSERACT_PATH.getParent(), "tessdata");
-        if (!languagePackRootDir.exists()) {
-            return "";
-        }
+        File languagePackRootDir = new File(PlatformUtil.getOcrLanguagePacksPath());
 
         List<String> languagePacks = new ArrayList<>();
         for (File languagePack : languagePackRootDir.listFiles()) {