add support for OCR ...

Providers/Extractors/XImageFileContentExtractor.cs

@@ -5,6 +5,10 @@ using System.Threading;
 using xAiModels.Models;
 using System.Threading.Tasks;
 using xAiApi.Interfaces.Extractors;
+using xAiApi.Configurations;
+using xCommons.Extensions;
+using xExceptions.Constants;
+using Tesseract;
 
 namespace xAiApi.Providers.Extractors
 {
@@ -24,16 +28,25 @@ namespace xAiApi.Providers.Extractors
         ];
 
         /// <summary>
-        /// آیا OCR فعال است (برای مدل‌های غیر Vision) ...
+        /// AI Api Configuration ...
         /// </summary>
-        private readonly bool _enableOcrFallback;
+        private readonly XAiApiConfiguration configuration;
 
-        public XImageFileContentExtractor() : this(enableOcrFallback: true)
-        { }
-
-        public XImageFileContentExtractor(bool enableOcrFallback = true)
+        public XImageFileContentExtractor(
+            XAiApiConfiguration configuration
+        )
         {
-            _enableOcrFallback = enableOcrFallback;
+            //
+            if (configuration.IsNullOrDefault() ||
+                configuration.OCR.IsNullOrDefault() ||
+                configuration.OCR.DataPath.IsNullOrEmpty() ||
+                configuration.OCR.DefaultLanguage.IsNullOrEmpty())
+            {
+                XException.InvalidConfiguration.Throw();
+            }
+
+            //
+            this.configuration = configuration;
         }
 
         /// <summary>
@@ -103,8 +116,9 @@ namespace xAiApi.Providers.Extractors
                 Description = $"Attached image: {fileName}"
             });
 
-            // OCR Fallback برای مدل‌های متنی
-            if (_enableOcrFallback)
+            //
+            // OCR Fallback ...
+            if (configuration.OCR.EnableOcrFallback)
             {
                 //
                 try
@@ -135,8 +149,44 @@ namespace xAiApi.Providers.Extractors
         )
         {
             //
-            // TODO: Complete this ...
-            return await Task.FromResult(string.Empty);
+            return await Task.Run(() =>
+            {
+                //
+                // Validate Path ...
+                if (!Directory.Exists(configuration.OCR.DataPath))
+                {
+                    //
+                    throw new DirectoryNotFoundException(
+                        $"TessData directory not found at: {configuration.OCR.DataPath}. " +
+                        "Please download traineddata files from https://github.com/tesseract-ocr/tessdata"
+                    );
+                }
+            
+                //
+                using var engine = new TesseractEngine(
+                    datapath: configuration.OCR.DataPath,
+                    language: configuration.OCR.DefaultLanguage,
+                    engineMode: configuration.OCR.EngineMode
+                );
+
+                //
+                using var pix = Pix.LoadFromMemory(imageBytes);
+
+                //
+                cancellationToken.ThrowIfCancellationRequested();
+
+                //
+                using var page = engine.Process(pix);
+
+                //
+                var text = page.GetText();
+
+                //
+                text = text?.Trim();
+
+                //
+                return text ?? string.Empty;
+            }, cancellationToken);
         }
     }
 }