Add support for workflow_ocr_backend

* Integration for https://github.com/R0Wi-DEV/workflow_ocr_backend
* Implements #51

lib/AppInfo/Application.php

@@ -33,8 +33,12 @@
 use OCA\WorkflowOcr\Helper\SidecarFileAccessor;
 use OCA\WorkflowOcr\Listener\RegisterFlowOperationsListener;
 use OCA\WorkflowOcr\Notification\Notifier;
+use OCA\WorkflowOcr\OcrProcessors\CommandLineUtils;
+use OCA\WorkflowOcr\OcrProcessors\ICommandLineUtils;
 use OCA\WorkflowOcr\OcrProcessors\IOcrProcessorFactory;
 use OCA\WorkflowOcr\OcrProcessors\OcrProcessorFactory;
+use OCA\WorkflowOcr\OcrProcessors\Remote\Client\ApiClient;
+use OCA\WorkflowOcr\OcrProcessors\Remote\Client\IApiClient;
 use OCA\WorkflowOcr\Service\EventService;
 use OCA\WorkflowOcr\Service\GlobalSettingsService;
 use OCA\WorkflowOcr\Service\IEventService;
@@ -46,8 +50,10 @@
 use OCA\WorkflowOcr\Service\OcrBackendInfoService;
 use OCA\WorkflowOcr\Service\OcrService;
 use OCA\WorkflowOcr\SetupChecks\OcrMyPdfCheck;
+use OCA\WorkflowOcr\Wrapper\AppApiWrapper;
 use OCA\WorkflowOcr\Wrapper\CommandWrapper;
 use OCA\WorkflowOcr\Wrapper\Filesystem;
+use OCA\WorkflowOcr\Wrapper\IAppApiWrapper;
 use OCA\WorkflowOcr\Wrapper\ICommand;
 use OCA\WorkflowOcr\Wrapper\IFilesystem;
 use OCA\WorkflowOcr\Wrapper\IViewFactory;
@@ -63,6 +69,8 @@
 
 class Application extends App implements IBootstrap {
 	public const APP_NAME = 'workflow_ocr';
+	public const APP_BACKEND_NAME = 'workflow_ocr_backend';
+	public const APP_API_APP_NAME = 'app_api';
 
 	/**
 	 * Application constructor.
@@ -83,6 +91,9 @@ public function register(IRegistrationContext $context): void {
 		$context->registerServiceAlias(IEventService::class, EventService::class);
 		$context->registerServiceAlias(IOcrBackendInfoService::class, OcrBackendInfoService::class);
 		$context->registerServiceAlias(INotificationService::class, NotificationService::class);
+		$context->registerServiceAlias(IApiClient::class, ApiClient::class);
+		$context->registerServiceAlias(ICommandLineUtils::class, CommandLineUtils::class);
+		$context->registerServiceAlias(IAppApiWrapper::class, AppApiWrapper::class);
 
 		// BUG #43
 		$context->registerService(ICommand::class, function () {

lib/Exception/OcrProcessorNotFoundException.php

@@ -26,7 +26,7 @@
 use Exception;
 
 class OcrProcessorNotFoundException extends Exception {
-	public function __construct(string $mimeType) {
-		$this->message = 'OCR processor for mime type ' . $mimeType . ' not found';
+	public function __construct(string $mimeType, bool $useRemoteBackend) {
+		$this->message = 'OCR processor for mime type ' . $mimeType . '(useRemoteBackend=' . $useRemoteBackend . ') not found';
 	}
 }

lib/OcrProcessors/CommandLineUtils.php

@@ -0,0 +1,100 @@
+<?php
+
+declare(strict_types=1);
+
+/**
+ * @copyright Copyright (c) 2025 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\OcrProcessors;
+
+use OCA\WorkflowOcr\Helper\ISidecarFileAccessor;
+use OCA\WorkflowOcr\Model\GlobalSettings;
+use OCA\WorkflowOcr\Model\WorkflowSettings;
+use OCA\WorkflowOcr\Service\IOcrBackendInfoService;
+use Psr\Log\LoggerInterface;
+
+class CommandLineUtils implements ICommandLineUtils {
+	private static $ocrModeToCmdParameterMapping = [
+		WorkflowSettings::OCR_MODE_SKIP_TEXT => '--skip-text',
+		WorkflowSettings::OCR_MODE_REDO_OCR => '--redo-ocr',
+		WorkflowSettings::OCR_MODE_FORCE_OCR => '--force-ocr',
+		WorkflowSettings::OCR_MODE_SKIP_FILE => '' // This is the ocrmypdf default behaviour
+	];
+
+	public function __construct(
+		private ISidecarFileAccessor $sidecarFileAccessor,
+		private IOcrBackendInfoService $ocrBackendInfoService,
+		private LoggerInterface $logger,
+	) {
+	}
+
+	public function getCommandlineArgs(WorkflowSettings $settings, GlobalSettings $globalSettings, array $additionalCommandlineArgs = []): string {
+		$isLocalExecution = !$this->ocrBackendInfoService->isRemoteBackend();
+
+		// Default setting is quiet
+		$args = $isLocalExecution ? ['-q'] : [];
+
+		// OCR mode ('--skip-text', '--redo-ocr', '--force-ocr' or empty)
+		$args[] = self::$ocrModeToCmdParameterMapping[$settings->getOcrMode()];
+
+		// Language settings
+		if ($settings->getLanguages()) {
+			$langStr = implode('+', $settings->getLanguages());
+			$args[] = "--language $langStr";
+		}
+
+		// Remove background option (NOTE :: this is incompatible with redo-ocr, so
+		// we have to make it exclusive against each other!)
+		if ($settings->getRemoveBackground()) {
+			if ($settings->getOcrMode() === WorkflowSettings::OCR_MODE_REDO_OCR) {
+				$this->logger->warning('--remove-background is incompatible with --redo-ocr, ignoring');
+			} else {
+				$args[] = '--remove-background';
+			}
+		}
+
+		// Number of CPU's to be used
+		$processorCount = intval($globalSettings->processorCount);
+		if ($processorCount > 0) {
+			$args[] = '--jobs ' . $processorCount;
+		}
+
+		if ($isLocalExecution) {
+			// Save recognized text in tempfile
+			$sidecarFilePath = $this->sidecarFileAccessor->getOrCreateSidecarFile();
+			if ($sidecarFilePath) {
+				$args[] = '--sidecar ' . $sidecarFilePath;
+			}
+		}
+
+		$resultArgs = array_filter(array_merge(
+			$args,
+			$additionalCommandlineArgs,
+			[$this->escapeCustomCliArgs($settings->getCustomCliArgs())]
+		), fn ($arg) => !empty($arg));
+
+		return implode(' ', $resultArgs);
+	}
+
+	private function escapeCustomCliArgs(string $customCliArgs): string {
+		$customCliArgs = str_replace('&&', '', $customCliArgs);
+		$customCliArgs = str_replace(';', '', $customCliArgs);
+		return $customCliArgs;
+	}
+}

lib/OcrProcessors/ICommandLineUtils.php

@@ -0,0 +1,31 @@
+<?php
+
+declare(strict_types=1);
+
+/**
+ * @copyright Copyright (c) 2025 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\OcrProcessors;
+
+use OCA\WorkflowOcr\Model\GlobalSettings;
+use OCA\WorkflowOcr\Model\WorkflowSettings;
+
+interface ICommandLineUtils {
+	public function getCommandlineArgs(WorkflowSettings $settings, GlobalSettings $globalSettings, array $additionalCommandlineArgs = []): string;
+}

lib/OcrProcessors/IOcrProcessorFactory.php

@@ -28,11 +28,4 @@ interface IOcrProcessorFactory {
 	 * Creates a IOcrProcessor object for the given mimetype
 	 */
 	public function create(string $mimeType) : IOcrProcessor;
-
-	/**
-	 * Returns true, if an OCR processor for the given mimetype
-	 * can be constructed.
-	 * @return bool
-	 */
-	public function canCreate(string $mimeType) : bool;
 }

lib/OcrProcessors/ImageOcrProcessor.php → lib/OcrProcessors/Local/ImageOcrProcessor.php

@@ -21,7 +21,7 @@
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
-namespace OCA\WorkflowOcr\OcrProcessors;
+namespace OCA\WorkflowOcr\OcrProcessors\Local;
 
 use OCA\WorkflowOcr\Model\GlobalSettings;
 use OCA\WorkflowOcr\Model\WorkflowSettings;

lib/OcrProcessors/OcrMyPdfBasedProcessor.php → lib/OcrProcessors/Local/OcrMyPdfBasedProcessor.php

@@ -21,42 +21,32 @@
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
-namespace OCA\WorkflowOcr\OcrProcessors;
+namespace OCA\WorkflowOcr\OcrProcessors\Local;
 
 use OCA\WorkflowOcr\Exception\OcrNotPossibleException;
 use OCA\WorkflowOcr\Exception\OcrResultEmptyException;
 use OCA\WorkflowOcr\Helper\ISidecarFileAccessor;
 use OCA\WorkflowOcr\Model\GlobalSettings;
 use OCA\WorkflowOcr\Model\WorkflowSettings;
+use OCA\WorkflowOcr\OcrProcessors\ICommandLineUtils;
+use OCA\WorkflowOcr\OcrProcessors\IOcrProcessor;
+use OCA\WorkflowOcr\OcrProcessors\OcrProcessorResult;
 use OCA\WorkflowOcr\Wrapper\ICommand;
 use OCP\Files\File;
 use Psr\Log\LoggerInterface;
 
 abstract class OcrMyPdfBasedProcessor implements IOcrProcessor {
-	private static $ocrModeToCmdParameterMapping = [
-		WorkflowSettings::OCR_MODE_SKIP_TEXT => '--skip-text',
-		WorkflowSettings::OCR_MODE_REDO_OCR => '--redo-ocr',
-		WorkflowSettings::OCR_MODE_FORCE_OCR => '--force-ocr',
-		WorkflowSettings::OCR_MODE_SKIP_FILE => '' // This is the ocrmypdf default behaviour
-	];
-
-	/** @var ICommand */
-	private $command;
-
-	/** @var LoggerInterface */
-	private $logger;
-
-	/** @var ISidecarFileAccessor */
-	private $sidecarFileAccessor;
-
-	public function __construct(ICommand $command, LoggerInterface $logger, ISidecarFileAccessor $sidecarFileAccessor) {
-		$this->command = $command;
-		$this->logger = $logger;
-		$this->sidecarFileAccessor = $sidecarFileAccessor;
+	public function __construct(
+		private ICommand $command,
+		private LoggerInterface $logger,
+		private ISidecarFileAccessor $sidecarFileAccessor,
+		private ICommandLineUtils $commandLineUtils,
+	) {
 	}
 
 	public function ocrFile(File $file, WorkflowSettings $settings, GlobalSettings $globalSettings): OcrProcessorResult {
-		$commandStr = 'ocrmypdf ' . $this->getCommandlineArgs($settings, $globalSettings) . ' - - || exit $? ; cat';
+		$additionalCommandlineArgs = $this->getAdditionalCommandlineArgs($settings, $globalSettings);
+		$commandStr = 'ocrmypdf ' . $this->commandLineUtils->getCommandlineArgs($settings, $globalSettings, $additionalCommandlineArgs) . ' - - || exit $? ; cat';
 
 		$inputFileContent = $file->getContent();
 
@@ -109,55 +99,4 @@ public function ocrFile(File $file, WorkflowSettings $settings, GlobalSettings $
 	protected function getAdditionalCommandlineArgs(WorkflowSettings $settings, GlobalSettings $globalSettings): array {
 		return [];
 	}
-
-
-	private function getCommandlineArgs(WorkflowSettings $settings, GlobalSettings $globalSettings): string {
-		// Default setting is quiet
-		$args = ['-q'];
-
-		// OCR mode ('--skip-text', '--redo-ocr', '--force-ocr' or empty)
-		$args[] = self::$ocrModeToCmdParameterMapping[$settings->getOcrMode()];
-
-		// Language settings
-		if ($settings->getLanguages()) {
-			$langStr = implode('+', $settings->getLanguages());
-			$args[] = "-l $langStr";
-		}
-
-		// Remove background option (NOTE :: this is incompatible with redo-ocr, so
-		// we have to make it exclusive against each other!)
-		if ($settings->getRemoveBackground()) {
-			if ($settings->getOcrMode() === WorkflowSettings::OCR_MODE_REDO_OCR) {
-				$this->logger->warning('--remove-background is incompatible with --redo-ocr, ignoring');
-			} else {
-				$args[] = '--remove-background';
-			}
-		}
-
-		// Number of CPU's to be used
-		$processorCount = intval($globalSettings->processorCount);
-		if ($processorCount > 0) {
-			$args[] = '-j ' . $processorCount;
-		}
-
-		// Save recognized text in tempfile
-		$sidecarFilePath = $this->sidecarFileAccessor->getOrCreateSidecarFile();
-		if ($sidecarFilePath) {
-			$args[] = '--sidecar ' . $sidecarFilePath;
-		}
-
-		$resultArgs = array_filter(array_merge(
-			$args,
-			$this->getAdditionalCommandlineArgs($settings, $globalSettings),
-			[$this->escapeCustomCliArgs($settings->getCustomCliArgs())]
-		), fn ($arg) => !empty($arg));
-
-		return implode(' ', $resultArgs);
-	}
-
-	private function escapeCustomCliArgs(string $customCliArgs): string {
-		$customCliArgs = str_replace('&&', '', $customCliArgs);
-		$customCliArgs = str_replace(';', '', $customCliArgs);
-		return $customCliArgs;
-	}
 }

lib/OcrProcessors/PdfOcrProcessor.php → lib/OcrProcessors/Local/PdfOcrProcessor.php

@@ -21,7 +21,7 @@
  * along with this program.  If not, see <http://www.gnu.org/licenses/>.
  */
 
-namespace OCA\WorkflowOcr\OcrProcessors;
+namespace OCA\WorkflowOcr\OcrProcessors\Local;
 
 class PdfOcrProcessor extends OcrMyPdfBasedProcessor {
 }