adding File Content Extraction Services ...

2026-10-01 01:16:24 +03:30
parent aac3eb92c1
commit f32972b55f
6 changed files with 225 additions and 2 deletions
@@ -0,0 +1,68 @@
+using System.IO;
+using System.Linq;
+using System.Threading;
+using xAiApi.Interfaces;
+using xCommons.Extensions;
+using xExceptions.Constants;
+using System.Threading.Tasks;
+using System.Collections.Generic;
+
+namespace xAiApi.Providers
+{
+    /// <summary>
+    /// Composite extractor that delegates to appropriate extractor 
+    /// based on MIME type ...
+    /// </summary>    
+    public class XCompositeFileContentExtractor : IXFileContentExtractor
+    {
+        //
+        private readonly IEnumerable<IXFileContentExtractor> extractors;
+
+        public XCompositeFileContentExtractor(
+            IEnumerable<IXFileContentExtractor> extractors
+        )
+        {
+            this.extractors = extractors;
+        }
+
+        /// <summary>
+        /// Check if this extractor supports the specified MIME type ...
+        /// </summary>
+        public bool CanExtract(string mimeType)
+        {
+            return extractors.Any(e => e.CanExtract(mimeType));
+        }
+
+        /// <summary>
+        /// Extract text content from file stream ...
+        /// </summary>
+        public async Task<string> ExtractAsync(
+            Stream fileStream,
+            string mimeType,
+            CancellationToken cancellationToken = default
+        )
+        {
+            //
+            var extractor = extractors
+                .FirstOrDefault(e => e.CanExtract(mimeType));
+            if (extractor.IsNull())
+            {
+                //
+                XException.NotAllowed.Throw(
+                    $"Unsupported file type: {mimeType}"
+                );
+            }
+
+            //
+            var result = await extractor
+                .ExtractAsync(
+                    mimeType: mimeType,
+                    fileStream: fileStream,
+                    cancellationToken: cancellationToken
+                );
+
+            //
+            return result;
+        }
+    }
+}