feat(kreuzberg): improve tool - remove preprocessing, better file search, show available files on error

- Removed image preprocessing (Kreuzberg handles it internally)
- Improved file search (partial name matching)
- Shows available files when file not found
- Uses mimetypes for proper content-type

effi-kreuzberg/TOOL-OPENWEBUI.md

@@ -1,10 +1,10 @@
-# Kreuzberg OCR Tool - Enhanced Version
+# Kreuzberg OCR Tool
 
 ## Installation
 
 1. **Admin Panel** → **Tools** → **Create Tool**
 2. **Name**: `Kreuzberg OCR`
-3. **Description**: `Extract text from documents using Kreuzberg OCR (PaddleOCR backend). Single: extract_file("file") or Multiple: extract_files("file1,file2")`
+3. **Description**: `Extract text from documents using Kreuzberg OCR. Usage: extract_file("filename") or extract_files("file1,file2")`
 4. **Coller le code ci-dessous**
 5. **Save and Activate**
 
@@ -16,7 +16,7 @@ import glob
 import requests
 import io
 import json
-from PIL import Image, ImageEnhance
+import mimetypes
 
 
 class Tools:
@@ -25,16 +25,6 @@ class Tools:
         self.uploads_dir = "/app/backend/data/uploads"
         self.host_uploads_dir = "/home/effi/effi/data/uploads"
 
-    def _preprocess_image(self, image):
-        if image.mode in ('RGBA', 'LA', 'P'):
-            image = image.convert('RGBA').convert('RGB')
-        elif image.mode != 'RGB':
-            image = image.convert('RGB')
-        image = image.convert('L')
-        enhancer = ImageEnhance.Contrast(image)
-        image = enhancer.enhance(2.0)
-        return image
-
     def extract_file(self, filename):
         return self._extract_single_file(filename)
 
@@ -76,23 +66,12 @@ class Tools:
             file_path = self._find_file(filename)
             
             if not file_path:
-                return "[ERROR] File not found: " + filename + ". Use list_recent_files() to see available files."
+                available = self._list_files()
+                return "[ERROR] File not found: " + filename + "\n\nAvailable files:\n" + available
             
-            mime_type = self._get_mime_type(file_path)
+            mime_type, _ = mimetypes.guess_type(file_path)
-            ext = os.path.splitext(file_path)[1].lower()
+            mime_type = mime_type or 'application/octet-stream'
-            is_image = ext in ('.jpg', '.jpeg', '.png', '.gif', '.bmp', '.tiff', '.tif', '.webp')
             
-            if is_image:
-                with open(file_path, 'rb') as f:
-                    img = Image.open(f)
-                    img = self._preprocess_image(img)
-                    img_bytes = io.BytesIO()
-                    img.save(img_bytes, format='PNG')
-                    img_bytes.seek(0)
-                files = {
-                    'files': (os.path.basename(file_path) + '.png', img_bytes, 'image/png')
-                }
-            else:
             with open(file_path, 'rb') as f:
                 files = {
                     'files': (os.path.basename(file_path), f, mime_type)
@@ -119,11 +98,11 @@ class Tools:
                         content = self._clean_content(content)
                         
                         if len(content) > 15000:
-                            content = content[:15000] + "\n\n... (document truncated)"
+                            content = content[:15000] + "\n\n... (truncated)"
                         
                         return "FILE: " + os.path.basename(file_path) + "\n\n" + content
                     else:
-                        return "[WARNING] No text extracted from " + os.path.basename(file_path) + ". File may be empty or contain only images."
+                        return "[WARNING] No text extracted from " + os.path.basename(file_path)
                 
                 return "[WARNING] Unexpected response from Kreuzberg"
             
@@ -134,46 +113,23 @@ class Tools:
                 return "[ERROR] Kreuzberg error (" + str(response.status_code) + ")"
         
         except requests.Timeout:
-            return "[ERROR] Timeout - file is too large or Kreuzberg is overloaded"
+            return "[ERROR] Timeout - file is too large"
-        
         except requests.ConnectionError:
             return "[ERROR] Cannot connect to Kreuzberg at " + self.kreuzberg_url
-        
         except Exception as e:
             return "[ERROR] " + str(e)
 
     def list_recent_files(self, count=10):
-        try:
+        available = self._list_files(count)
-            files = glob.glob(os.path.join(self.uploads_dir, "*"))
+        return available
-            
-            if not files and os.path.exists(self.host_uploads_dir):
-                files = glob.glob(os.path.join(self.host_uploads_dir, "*"))
-            
-            files = [f for f in files if os.path.isfile(f)]
-            files = sorted(files, key=os.path.getmtime, reverse=True)[:count]
-            
-            if not files:
-                return "[INFO] No files found in uploads directory."
-            
-            result = "[INFO] " + str(len(files)) + " recent files:\n\n"
-            for f in files:
-                name = os.path.basename(f)
-                size = os.path.getsize(f)
-                size_str = self._format_size(size)
-                result += "- " + name + " (" + size_str + ")\n"
-            
-            result += "\nUse extract_file(\"filename\") to OCR a file."
-            return result
-        
-        except Exception as e:
-            return "[ERROR] " + str(e)
 
     def search_files(self, pattern):
         try:
-            files = glob.glob(os.path.join(self.uploads_dir, "*" + pattern + "*"))
+            search_dirs = [d for d in [self.uploads_dir, self.host_uploads_dir] if os.path.exists(d)]
             
-            if not files and os.path.exists(self.host_uploads_dir):
+            files = []
-                files = glob.glob(os.path.join(self.host_uploads_dir, "*" + pattern + "*"))
+            for directory in search_dirs:
+                files.extend(glob.glob(os.path.join(directory, "*" + pattern + "*")))
             
             files = [f for f in files if os.path.isfile(f)]
             
@@ -191,64 +147,60 @@ class Tools:
         except Exception as e:
             return "[ERROR] " + str(e)
 
+    def _list_files(self, count=10):
+        try:
+            files = []
+            search_dirs = [d for d in [self.uploads_dir, self.host_uploads_dir] if os.path.exists(d)]
+            
+            for directory in search_dirs:
+                files.extend(glob.glob(os.path.join(directory, "*")))
+            
+            files = [f for f in files if os.path.isfile(f)]
+            files.sort(key=os.path.getmtime, reverse=True)
+            
+            if not files:
+                return "[INFO] No files found in uploads directory."
+            
+            result = "[INFO] " + str(len(files)) + " recent files:\n\n"
+            for f in files[:count]:
+                name = os.path.basename(f)
+                size = os.path.getsize(f)
+                size_str = self._format_size(size)
+                result += "- " + name + " (" + size_str + ")\n"
+            
+            result += "\nUse extract_file(\"filename\") to OCR a file."
+            return result
+        
+        except Exception as e:
+            return "[ERROR] " + str(e)
+
     def _find_file(self, filename):
         filename = filename.strip().strip('"').strip("'")
         
         if os.path.exists(filename):
             return filename
         
-        path = os.path.join(self.uploads_dir, filename)
+        search_dirs = [d for d in [self.uploads_dir, self.host_uploads_dir] if os.path.exists(d)]
+        
+        for directory in search_dirs:
+            path = os.path.join(directory, filename)
             if os.path.exists(path):
                 return path
         
-        path = os.path.join(self.host_uploads_dir, filename)
+        for directory in search_dirs:
-        if os.path.exists(path):
+            all_files = glob.glob(os.path.join(directory, "*"))
-            return path
+            for f in all_files:
-        
+                basename = os.path.basename(f).lower()
-        for directory in [self.uploads_dir, self.host_uploads_dir]:
+                search_name = filename.lower()
-            if not os.path.exists(directory):
+                if search_name in basename or basename.startswith(search_name):
-                continue
+                    return f
-            
+                parts = search_name.replace('.jpeg', '').replace('.jpg', '').replace('.png', '').split()
-            pattern = os.path.join(directory, filename + "*")
+                for part in parts:
-            matches = glob.glob(pattern)
+                    if len(part) > 3 and part in basename:
-            if matches:
+                        return f
-                return matches[0]
-            
-            pattern = os.path.join(directory, "*" + filename + "*")
-            matches = glob.glob(pattern)
-            if matches:
-                for m in matches:
-                    if filename in os.path.basename(m):
-                        return m
-                return matches[0]
         
         return None
 
-    def _get_mime_type(self, file_path):
-        ext = os.path.splitext(file_path)[1].lower()
-        mime_types = {
-            '.pdf': 'application/pdf',
-            '.jpg': 'image/jpeg',
-            '.jpeg': 'image/jpeg',
-            '.png': 'image/png',
-            '.gif': 'image/gif',
-            '.tiff': 'image/tiff',
-            '.tif': 'image/tiff',
-            '.bmp': 'image/bmp',
-            '.webp': 'image/webp',
-            '.doc': 'application/msword',
-            '.docx': 'application/vnd.openxmlformats-officedocument.wordprocessingml.document',
-            '.xls': 'application/vnd.ms-excel',
-            '.xlsx': 'application/vnd.openxmlformats-officedocument.spreadsheetml.sheet',
-            '.ppt': 'application/vnd.ms-powerpoint',
-            '.pptx': 'application/vnd.openxmlformats-officedocument.presentationml.presentation',
-            '.epub': 'application/epub+zip',
-            '.txt': 'text/plain',
-            '.html': 'text/html',
-            '.htm': 'text/html',
-        }
-        return mime_types.get(ext, 'application/octet-stream')
-
     def _clean_content(self, content):
         if not content:
             return ""
@@ -267,23 +219,15 @@ class Tools:
                 cleaned_lines.append(stripped)
                 prev_empty = False
         
-        result = '\n'.join(cleaned_lines)
+        return '\n'.join(cleaned_lines)
-        
-        if len(result) > 500:
-            short_words = sum(1 for w in result.split() if len(w) <= 3)
-            total_words = len(result.split())
-            if total_words > 50 and short_words / total_words > 0.3:
-                result += "\n\n[WARNING] This text may contain OCR errors."
-        
-        return result
 
     def _format_size(self, size):
         if size < 1024:
             return str(size) + " B"
         elif size < 1024 * 1024:
-            return str(size / 1024) + " KB"
+            return str(int(size / 1024)) + " KB"
         else:
-            return str(size / (1024 * 1024)) + " MB"
+            return str(int(size / (1024 * 1024))) + " MB"
 
 
 def extract_file(filename):
@@ -305,6 +249,20 @@ def search_files(pattern):
 
 ## Usage Examples
 
+```
+extract_file("document.pdf")
+extract_file("WhatsApp Image 2026-04-05 at 10.20.56.jpeg")
+extract_files("page1.jpg,page2.jpg,page3.jpg")
+list_recent_files()
+search_files("dnd")
+```
+
+---
+
+*Updated 11 April 2026 - Improved file search, no preprocessing (Kreuzberg handles it)*
+
+## Usage Examples
+
 After installing and activating the tool:
 
 ### Extract text from a single file: