Add Next.js frontend with WebLLM, OpenAI support - Add complete Next.js frontend with Tailwind CSS and shadcn/ui - Integrate WebLLM for client-side browser-based translations - Add OpenAI provider support with gpt-4o-mini default - Add Context & Glossary page for LLM customization - Reorganize settings: Translation Services includes all providers - Add system prompt and glossary support for all LLMs - Remove test files and requirements-test.txt

services/translation_service.py

@@ -54,15 +54,19 @@ class DeepLTranslationProvider(TranslationProvider):
 class LibreTranslationProvider(TranslationProvider):
     """LibreTranslate implementation"""
     
+    def __init__(self, custom_url: str = "https://libretranslate.com"):
+        self.custom_url = custom_url
+    
     def translate(self, text: str, target_language: str, source_language: str = 'auto') -> str:
         if not text or not text.strip():
             return text
         
         try:
-            # LibreTranslator doesn't need API key for self-hosted instances
-            translator = LibreTranslator(source=source_language, target=target_language, custom_url="http://localhost:5000")
+            # LibreTranslator supports custom URL for self-hosted or public instances
+            translator = LibreTranslator(source=source_language, target=target_language, custom_url=self.custom_url)
             return translator.translate(text)
         except Exception as e:
+            print(f"LibreTranslate error: {e}")
             # Fail silently and return original text
             return text
 
@@ -188,6 +192,97 @@ class WebLLMTranslationProvider(TranslationProvider):
         return text
 
 
+class OpenAITranslationProvider(TranslationProvider):
+    """OpenAI GPT translation implementation with vision support"""
+    
+    def __init__(self, api_key: str, model: str = "gpt-4o-mini", system_prompt: str = ""):
+        self.api_key = api_key
+        self.model = model
+        self.custom_system_prompt = system_prompt
+    
+    def translate(self, text: str, target_language: str, source_language: str = 'auto') -> str:
+        if not text or not text.strip():
+            return text
+        
+        # Skip very short text or numbers only
+        if len(text.strip()) < 2 or text.strip().isdigit():
+            return text
+        
+        try:
+            import openai
+            client = openai.OpenAI(api_key=self.api_key)
+            
+            # Build system prompt with custom context if provided
+            base_prompt = f"You are a translator. Translate the user's text to {target_language}. Return ONLY the translation, nothing else."
+            
+            if self.custom_system_prompt:
+                system_content = f"""{base_prompt}
+
+ADDITIONAL CONTEXT AND INSTRUCTIONS:
+{self.custom_system_prompt}"""
+            else:
+                system_content = base_prompt
+            
+            response = client.chat.completions.create(
+                model=self.model,
+                messages=[
+                    {"role": "system", "content": system_content},
+                    {"role": "user", "content": text}
+                ],
+                temperature=0.3,
+                max_tokens=500
+            )
+            
+            translated = response.choices[0].message.content.strip()
+            return translated if translated else text
+        except Exception as e:
+            print(f"OpenAI translation error: {e}")
+            return text
+    
+    def translate_image(self, image_path: str, target_language: str) -> str:
+        """Translate text within an image using OpenAI vision model"""
+        import base64
+        
+        try:
+            import openai
+            client = openai.OpenAI(api_key=self.api_key)
+            
+            # Read and encode image
+            with open(image_path, 'rb') as img_file:
+                image_data = base64.b64encode(img_file.read()).decode('utf-8')
+            
+            # Determine image type from extension
+            ext = image_path.lower().split('.')[-1]
+            media_type = f"image/{ext}" if ext in ['png', 'jpg', 'jpeg', 'gif', 'webp'] else "image/png"
+            
+            response = client.chat.completions.create(
+                model=self.model,  # gpt-4o and gpt-4o-mini support vision
+                messages=[
+                    {
+                        "role": "user",
+                        "content": [
+                            {
+                                "type": "text",
+                                "text": f"Extract all text from this image and translate it to {target_language}. Return ONLY the translated text, preserving the structure and formatting."
+                            },
+                            {
+                                "type": "image_url",
+                                "image_url": {
+                                    "url": f"data:{media_type};base64,{image_data}"
+                                }
+                            }
+                        ]
+                    }
+                ],
+                max_tokens=1000
+            )
+            
+            return response.choices[0].message.content.strip()
+        except Exception as e:
+            print(f"OpenAI vision translation error: {e}")
+            return ""
+
+
 class TranslationService:
     """Main translation service that delegates to the configured provider"""
     
@@ -224,7 +319,7 @@ class TranslationService:
     
     def translate_image(self, image_path: str, target_language: str) -> str:
         """
-        Translate text in an image using vision model (Ollama only)
+        Translate text in an image using vision model (Ollama or OpenAI)
         
         Args:
             image_path: Path to image file
@@ -236,9 +331,11 @@ class TranslationService:
         if not self.translate_images:
             return ""
         
-        # Only Ollama supports image translation
+        # Ollama and OpenAI support image translation
         if isinstance(self.provider, OllamaTranslationProvider):
             return self.provider.translate_image(image_path, target_language)
+        elif isinstance(self.provider, OpenAITranslationProvider):
+            return self.provider.translate_image(image_path, target_language)
         
         return ""