Implement #140

Get installed tesseract languages from backend

Signed-off-by: Robin Windey <[email protected]>

.github/workflows/build.yml

@@ -1,6 +1,7 @@
 name: Build artifact
 
 on:
+  pull_request:
   workflow_dispatch:
 
 env:
@@ -36,5 +37,5 @@ jobs:
       - name: Upload artifacts
         uses: actions/upload-artifact@v1
         with:
-          name: ${{ env.APP_NAME }}.tar.gz 
-          path: ${{ env.APP_NAME }}/build/artifacts/appstore/${{ env.APP_NAME }}.tar.gz      
+          name: ${{ env.APP_NAME }}.tar.gz
+          path: ${{ env.APP_NAME }}/build/artifacts/appstore/${{ env.APP_NAME }}.tar.gz

Makefile

@@ -180,6 +180,7 @@ appstore:
 	--exclude="../$(app_name)/*.json" \
 	--exclude="../$(app_name)/*.lock" \
 	--exclude="../$(app_name)/*.cov" \
+	--exclude="../$(app_name)/psalm.xml" \
 	../$(app_name) \
 
 .PHONY: test

appinfo/routes.php

@@ -27,7 +27,8 @@
 
 return [
 	'routes' => [
-		['name' => 'GlobalSettings#getGlobalSettings', 'url' => '/globalsettings', 'verb' => 'GET'],
-		['name' => 'GlobalSettings#setGlobalSettings', 'url' => '/globalsettings', 'verb' => 'PUT']
+		['name' => 'GlobalSettings#getGlobalSettings', 'url' => '/globalSettings', 'verb' => 'GET'],
+		['name' => 'GlobalSettings#setGlobalSettings', 'url' => '/globalSettings', 'verb' => 'PUT'],
+		['name' => 'OcrBackendInfo#getInstalledLanguages', 'url' => '/ocrBackendInfo/installedLangs', 'verb' => 'GET']
 	]
 ];

lib/AppInfo/Application.php

@@ -38,7 +38,9 @@
 use OCA\WorkflowOcr\Service\EventService;
 use OCA\WorkflowOcr\Service\GlobalSettingsService;
 use OCA\WorkflowOcr\Service\IGlobalSettingsService;
+use OCA\WorkflowOcr\Service\IOcrBackendInfoService;
 use OCA\WorkflowOcr\Service\IOcrService;
+use OCA\WorkflowOcr\Service\OcrBackendInfoService;
 use OCA\WorkflowOcr\Service\OcrService;
 use OCA\WorkflowOcr\Wrapper\CommandWrapper;
 use OCA\WorkflowOcr\Wrapper\Filesystem;
@@ -76,6 +78,7 @@ public function register(IRegistrationContext $context): void {
 		$context->registerServiceAlias(IFilesystem::class, Filesystem::class);
 		$context->registerServiceAlias(IGlobalSettingsService::class, GlobalSettingsService::class);
 		$context->registerServiceAlias(IEventService::class, EventService::class);
+		$context->registerServiceAlias(IOcrBackendInfoService::class, OcrBackendInfoService::class);
 
 		// BUG #43
 		$context->registerService(ICommand::class, function () {

lib/Controller/ControllerBase.php

@@ -0,0 +1,41 @@
+<?php
+
+declare(strict_types=1);
+
+/**
+ * @copyright Copyright (c) 2022 Robin Windey <[email protected]>
+ *
+ * @author Robin Windey <[email protected]>
+ *
+ * @license GNU AGPL version 3 or any later version
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as
+ * published by the Free Software Foundation, either version 3 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+
+namespace OCA\WorkflowOcr\Controller;
+
+use OCP\AppFramework\Controller;
+use OCP\AppFramework\Http\JSONResponse;
+
+abstract class ControllerBase extends Controller {
+	protected function tryExecute(callable $function) : JSONResponse {
+		try {
+			$result = $function();
+			return new JSONResponse($result);
+		} catch (\Throwable $e) {
+			return new JSONResponse(['error' => $e->getMessage()], 500);
+		}
+	}
+}

lib/Controller/GlobalSettingsController.php

@@ -28,14 +28,13 @@
 
 use OCA\WorkflowOcr\Model\GlobalSettings;
 use OCA\WorkflowOcr\Service\IGlobalSettingsService;
-use OCP\AppFramework\Controller;
 use OCP\AppFramework\Http\JSONResponse;
 use OCP\IRequest;
 
 /**
  * This is the backend API controller for the Admin.vue component.
  */
-class GlobalSettingsController extends Controller {
+class GlobalSettingsController extends ControllerBase {
 	/** @var IGlobalSettingsService */
 	private $globalSettingsService;
 
@@ -66,13 +65,4 @@ public function setGlobalSettings(array $globalSettings) : JSONResponse {
 			return $this->globalSettingsService->getGlobalSettings();
 		});
 	}
-
-	private function tryExecute(callable $function) : JSONResponse {
-		try {
-			$result = $function();
-			return new JSONResponse($result);
-		} catch (\Throwable $e) {
-			return new JSONResponse(['error' => $e->getMessage()], 500);
-		}
-	}
 }

lib/Controller/OcrBackendInfoController.php

@@ -0,0 +1,53 @@
+<?php
+
+declare(strict_types=1);
+
+/**
+ * @copyright Copyright (c) 2022 Robin Windey <[email protected]>
+ *
+ * @author Robin Windey <[email protected]>
+ *
+ * @license GNU AGPL version 3 or any later version
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as
+ * published by the Free Software Foundation, either version 3 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+
+namespace OCA\WorkflowOcr\Controller;
+
+use OCA\WorkflowOcr\Service\IOcrBackendInfoService;
+use OCP\AppFramework\Http\JSONResponse;
+use OCP\IRequest;
+
+/**
+ * This is the backend API controller which provides informations about the OCR backend system.
+ */
+class OcrBackendInfoController extends ControllerBase {
+	/** @var IOcrBackendInfoService */
+	private $ocrBackendInfoService;
+
+	public function __construct($AppName, IRequest $request, IOcrBackendInfoService $ocrBackendInfoService) {
+		parent::__construct($AppName, $request);
+		$this->ocrBackendInfoService = $ocrBackendInfoService;
+	}
+
+	/**
+	 * @return JSONResponse
+	 */
+	public function getInstalledLanguages() : JSONResponse {
+		return $this->tryExecute(function () {
+			return $this->ocrBackendInfoService->getInstalledLanguages();
+		});
+	}
+}

lib/Migration/Version2404Date20220903071748.php

@@ -0,0 +1,142 @@
+<?php
+
+declare(strict_types=1);
+
+/**
+ * @copyright Copyright (c) 2022 Robin Windey <[email protected]>
+ *
+ *  @license GNU AGPL version 3 or any later version
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as
+ * published by the Free Software Foundation, either version 3 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+
+namespace OCA\WorkflowOcr\Migration;
+
+use Closure;
+use Exception;
+use OCP\DB\ISchemaWrapper;
+use OCP\IDBConnection;
+use OCP\Migration\IOutput;
+use OCP\Migration\SimpleMigrationStep;
+
+class Version2404Date20220903071748 extends SimpleMigrationStep {
+
+	/** @var IDBConnection */
+	private $db;
+
+	public function __construct(IDBConnection $db) {
+		$this->db = $db;
+	}
+
+	/**
+	 * {@inheritDoc}
+	 */
+	public function name(): string {
+		return 'migrate lang codes';
+	}
+
+	/**
+	 * {@inheritDoc}
+	 */
+	public function description(): string {
+		return 'Execute migration of language codes towards tesseract langugage codes (e.g. deu instead of de)';
+	}
+
+	/**
+	 * {@inheritDoc}
+	 */
+	public function changeSchema(IOutput $output, Closure $schemaClosure, array $options): ?ISchemaWrapper {
+		// 'id' and new 'operation' value will be stored here
+		$datasetsToMigrate = $this->getDatasetsToMigrate();
+		$this->updateDatabase($datasetsToMigrate);
+
+		return null;
+	}
+
+	private function getDatasetsToMigrate() : array {
+		$langMapping = [
+			'de' => 'deu',
+			'en' => 'eng',
+			'fr' => 'fra',
+			'it' => 'ita',
+			'es' => 'spa',
+			'pt' => 'por',
+			'ru' => 'rus',
+			'chi' => 'chi_sim'
+		];
+
+		$builder = $this->db->getQueryBuilder();
+
+		$ocrFlowOperations = $builder->select('id', 'operation')
+			->from('flow_operations')
+			->where($builder->expr()->eq('class', $builder->createNamedParameter('OCA\WorkflowOcr\Operation')))
+			->executeQuery();
+
+		$datasetsToMigrate = [];
+
+		try {
+			while ($row = $ocrFlowOperations->fetch()) {
+				$workflowSettings = json_decode($row['operation'], true);
+				$foundMapping = false;
+				$newLangArr = [];
+				$languagesArr = $workflowSettings['languages'];
+
+				// Check if we need to migrate the languages code.
+				// If yes, we have to regenerate the whole 'operation' string.
+				foreach ($languagesArr as $existingLang) {
+					if (array_key_exists($existingLang, $langMapping)) {
+						$newLangArr[] = $langMapping[$existingLang];
+						$foundMapping = true;
+						continue;
+					}
+					$newLangArr[] = $existingLang;
+				}
+
+				if ($foundMapping) {
+					$workflowSettings['languages'] = $newLangArr;
+					$datasetsToMigrate[] = [
+						'id' => $row['id'],
+						'operation' => json_encode($workflowSettings)
+					];
+				}
+			}
+		} finally {
+			$ocrFlowOperations->closeCursor();
+		}
+
+		return $datasetsToMigrate;
+	}
+
+	private function updateDatabase(array $datasetsToMigrate) : void {
+		$this->db->beginTransaction();
+
+		try {
+			$builder = $this->db->getQueryBuilder();
+			$builder->update('flow_operations')
+				->set('operation', $builder->createParameter('operation'))
+				->where($builder->expr()->eq('id', $builder->createParameter('id')));
+
+			foreach ($datasetsToMigrate as $dataset) {
+				$builder->setParameter('id', $dataset['id']);
+				$builder->setParameter('operation', $dataset['operation']);
+				$builder->executeStatement();
+			}
+		} catch (Exception $e) {
+			$this->db->rollBack();
+			throw $e;
+		}
+
+		$this->db->commit();
+	}
+}

lib/OcrProcessors/OcrMyPdfBasedProcessor.php

@@ -33,22 +33,6 @@
 use Psr\Log\LoggerInterface;
 
 abstract class OcrMyPdfBasedProcessor implements IOcrProcessor {
-	/** @var array
-	 * Mapping for VUE frontend lang settings.
-	 * See also https://github.com/tesseract-ocr/tesseract/blob/main/doc/tesseract.1.asc#languages
-	 */
-	private static $langMapping = [
-		'de' => 'deu',
-		'en' => 'eng',
-		'fr' => 'fra',
-		'it' => 'ita',
-		'es' => 'spa',
-		'pt' => 'por',
-		'ru' => 'rus',
-		'chi' => 'chi_sim',
-		'est' => 'est',
-		'slk' => 'slk'
-	];
 
 	/** @var ICommand */
 	private $command;
@@ -127,14 +111,7 @@ private function getCommandlineArgs(WorkflowSettings $settings, GlobalSettings $
 
 		// Language settings
 		if ($settings->getLanguages()) {
-			$langStr = Chain::create($settings->getLanguages())
-				->map(function ($langCode) {
-					return self::$langMapping[(string)$langCode] ?? null;
-				})
-				->filter(function ($l) {
-					return $l !== null;
-				})
-				->join('+');
+			$langStr = Chain::create($settings->getLanguages())->join('+');
 			$args[] = "-l $langStr";
 		}
 

lib/Service/IOcrBackendInfoService.php

@@ -0,0 +1,38 @@
+<?php
+
+declare(strict_types=1);
+
+/**
+ * @copyright Copyright (c) 2022 Robin Windey <[email protected]>
+ *
+ * @author Robin Windey <[email protected]>
+ *
+ * @license GNU AGPL version 3 or any later version
+ *
+ * This program is free software: you can redistribute it and/or modify
+ * it under the terms of the GNU Affero General Public License as
+ * published by the Free Software Foundation, either version 3 of the
+ * License, or (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU Affero General Public License for more details.
+ *
+ * You should have received a copy of the GNU Affero General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ *
+ */
+
+namespace OCA\WorkflowOcr\Service;
+
+interface IOcrBackendInfoService {
+
+	/**
+	 * Returns all languages that are supported by the OCR backend.
+	 * Languages will be returned as an array of language-code-strings,
+	 * currently defined at https://github.com/tesseract-ocr/tesseract/blob/main/doc/tesseract.1.asc#languages.
+	 * @return array string[]
+	 */
+	public function getInstalledLanguages() : array;
+}