feat: add model existence check for Ollama integration (#227)

- Introduced a new function `check_ollama_model_exists` to verify the availability of specified Ollama models before usage.
- Added a custom exception `OllamaModelNotFoundError` for better error handling.
- Updated the RAG component to raise an exception if the required Ollama model is not found.
- Enhanced error handling in the UI to display specific messages related to embedding model issues.

Author: mmlmt2604

api/ollama_patch.py

@@ -5,13 +5,60 @@
 import adalflow as adal
 from adalflow.core.types import Document
 from adalflow.core.component import DataComponent
+import requests
+import os
 
 # Configure logging
 from api.logging_config import setup_logging
 
 setup_logging()
 logger = logging.getLogger(__name__)
 
+class OllamaModelNotFoundError(Exception):
+    """Custom exception for when Ollama model is not found"""
+    pass
+
+def check_ollama_model_exists(model_name: str, ollama_host: str = None) -> bool:
+    """
+    Check if an Ollama model exists before attempting to use it.
+    
+    Args:
+        model_name: Name of the model to check
+        ollama_host: Ollama host URL, defaults to localhost:11434
+        
+    Returns:
+        bool: True if model exists, False otherwise
+    """
+    if ollama_host is None:
+        ollama_host = os.getenv("OLLAMA_HOST", "http://localhost:11434")
+    
+    try:
+        # Remove /api prefix if present and add it back
+        if ollama_host.endswith('/api'):
+            ollama_host = ollama_host[:-4]
+        
+        response = requests.get(f"{ollama_host}/api/tags", timeout=5)
+        if response.status_code == 200:
+            models_data = response.json()
+            available_models = [model.get('name', '').split(':')[0] for model in models_data.get('models', [])]
+            model_base_name = model_name.split(':')[0]  # Remove tag if present
+            
+            is_available = model_base_name in available_models
+            if is_available:
+                logger.info(f"Ollama model '{model_name}' is available")
+            else:
+                logger.warning(f"Ollama model '{model_name}' is not available. Available models: {available_models}")
+            return is_available
+        else:
+            logger.warning(f"Could not check Ollama models, status code: {response.status_code}")
+            return False
+    except requests.exceptions.RequestException as e:
+        logger.warning(f"Could not connect to Ollama to check models: {e}")
+        return False
+    except Exception as e:
+        logger.warning(f"Error checking Ollama model availability: {e}")
+        return False
+
 class OllamaDocumentProcessor(DataComponent):
     """
     Process documents for Ollama embeddings by processing one document at a time.

api/rag.py

@@ -227,6 +227,17 @@ def __init__(self, provider="google", model=None, use_s3: bool = False):  # noqa
         # Determine if we're using Ollama embedder based on configuration
         self.is_ollama_embedder = is_ollama_embedder()
 
+        # Check if Ollama model exists before proceeding
+        if self.is_ollama_embedder:
+            from api.ollama_patch import check_ollama_model_exists
+            from api.config import get_embedder_config
+            
+            embedder_config = get_embedder_config()
+            if embedder_config and embedder_config.get("model_kwargs", {}).get("model"):
+                model_name = embedder_config["model_kwargs"]["model"]
+                if not check_ollama_model_exists(model_name):
+                    raise Exception(f"Ollama model '{model_name}' not found. Please run 'ollama pull {model_name}' to install it.")
+
         # Initialize components
         self.memory = Memory()
         self.embedder = get_embedder()

src/app/[owner]/[repo]/page.tsx

@@ -214,6 +214,7 @@ export default function RepoWikiPage() {
   const [requestInProgress, setRequestInProgress] = useState(false);
   const [currentToken, setCurrentToken] = useState(token); // Track current effective token
   const [effectiveRepoInfo, setEffectiveRepoInfo] = useState(repoInfo); // Track effective repo info with cached data
+  const [embeddingError, setEmbeddingError] = useState(false);
 
   // Model selection state variables
   const [selectedProviderState, setSelectedProviderState] = useState(providerParam);
@@ -597,6 +598,7 @@ Remember:
     if (!owner || !repo) {
       setError('Invalid repository information. Owner and repo name are required.');
       setIsLoading(false);
+      setEmbeddingError(false); // Reset embedding error state
       return;
     }
 
@@ -836,7 +838,17 @@ IMPORTANT:
         }
       }
 
-      // Clean up markdown delimiters
+      if(responseText.includes('Error preparing retriever: Environment variable OPENAI_API_KEY must be set')) {
+         setEmbeddingError(true);
+         throw new Error('OPENAI_API_KEY environment variable is not set. Please configure your OpenAI API key.');
+       }
+
+       if(responseText.includes('Ollama model') && responseText.includes('not found')) {
+         setEmbeddingError(true);
+         throw new Error('The specified Ollama embedding model was not found. Please ensure the model is installed locally or select a different embedding model in the configuration.');
+       }
+
+        // Clean up markdown delimiters
       responseText = responseText.replace(/^```(?:xml)?\s*/i, '').replace(/```\s*$/i, '');
 
       // Extract wiki structure from response
@@ -1074,6 +1086,7 @@ IMPORTANT:
     setGeneratedPages({});
     setPagesInProgress(new Set());
     setError(null);
+    setEmbeddingError(false); // Reset embedding error state
 
     try {
       // Set the request in progress flag
@@ -1471,6 +1484,7 @@ IMPORTANT:
     } catch (err) {
       console.warn('Error calling DELETE /api/wiki_cache:', err);
       setIsLoading(false);
+      setEmbeddingError(false); // Reset embedding error state
       // Optionally, inform the user about the cache clear error
       // setError(\`Error clearing cache: ${err instanceof Error ? err.message : String(err)}. Trying to refresh...\`);
       throw err;
@@ -1503,6 +1517,7 @@ IMPORTANT:
     setGeneratedPages({});
     setPagesInProgress(new Set());
     setError(null);
+    setEmbeddingError(false); // Reset embedding error state
     setIsLoading(true); // Set loading state for refresh
     setLoadingMessage(messages.loading?.initializing || 'Initializing wiki generation...');
 
@@ -1680,6 +1695,7 @@ IMPORTANT:
               setGeneratedPages(cachedData.generated_pages);
               setCurrentPageId(cachedStructure.pages.length > 0 ? cachedStructure.pages[0].id : undefined);
               setIsLoading(false);
+              setEmbeddingError(false); 
               setLoadingMessage(undefined);
               cacheLoadedSuccessfully.current = true;
               return; // Exit if cache is successfully loaded
@@ -1858,7 +1874,11 @@ IMPORTANT:
             </div>
             <p className="text-[var(--foreground)] text-sm mb-3">{error}</p>
             <p className="text-[var(--muted)] text-xs">
-              {messages.repoPage?.errorMessageDefault || 'Please check that your repository exists and is public. Valid formats are "owner/repo", "https://github.com/owner/repo", "https://gitlab.com/owner/repo", "https://bitbucket.org/owner/repo", or local folder paths like "C:\\path\\to\\folder" or "/path/to/folder".'}
+              {embeddingError ? (
+                messages.repoPage?.embeddingErrorDefault || 'This error is related to the document embedding system used for analyzing your repository. Please verify your embedding model configuration, API keys, and try again. If the issue persists, consider switching to a different embedding provider in the model settings.'
+              ) : (
+                messages.repoPage?.errorMessageDefault || 'Please check that your repository exists and is public. Valid formats are "owner/repo", "https://github.com/owner/repo", "https://gitlab.com/owner/repo", "https://bitbucket.org/owner/repo", or local folder paths like "C:\\path\\to\\folder" or "/path/to/folder".'
+              )}
             </p>
             <div className="mt-5">
               <Link

src/messages/en.json

@@ -112,6 +112,7 @@
     "home": "Home",
     "errorTitle": "Error",
     "errorMessageDefault": "Please check that your repository exists and is public. Valid formats are \"owner/repo\", \"https://github.com/owner/repo\", \"https://gitlab.com/owner/repo\", \"https://bitbucket.org/owner/repo\", or local folder paths like \"C:\\\\path\\\\to\\\\folder\" or \"/path/to/folder\".",
+    "embeddingErrorDefault": "This error is related to the document embedding system used for analyzing your repository. Please verify your embedding model configuration, API keys, and try again. If the issue persists, consider switching to a different embedding provider in the model settings.",
     "backToHome": "Back to Home",
     "exportWiki": "Export Wiki",
     "exportAsMarkdown": "Export as Markdown",

src/messages/es.json

@@ -103,6 +103,7 @@
     "home": "Inicio",
     "errorTitle": "Error",
     "errorMessageDefault": "Por favor, compruebe que su repositorio existe y es público. Los formatos válidos son \"owner/repo\", \"https://github.com/owner/repo\", \"https://gitlab.com/owner/repo\", \"https://bitbucket.org/owner/repo\", o rutas de carpetas locales como \"C:\\\\path\\\\to\\\\folder\" o \"/path/to/folder\".",
+    "embeddingErrorDefault": "Este error está relacionado con el sistema de embebido utilizado para analizar su repositorio. Por favor, verifique la configuración del modelo de embebido, las claves de API y vuelva a intentarlo. Si el problema persiste, considere cambiar al proveedor de embebido diferente en la configuración del modelo.",
     "backToHome": "Volver al Inicio",
     "exportWiki": "Exportar Wiki",
     "exportAsMarkdown": "Exportar como Markdown",

src/messages/ja.json

@@ -103,6 +103,7 @@
     "home": "ホーム",
     "errorTitle": "エラー",
     "errorMessageDefault": "リポジトリが存在し、公開されていることを確認してください。有効な形式は「owner/repo」、「https://github.com/owner/repo」、「https://gitlab.com/owner/repo」、「https://bitbucket.org/owner/repo」、またはローカルフォルダパス（例: 「C:\\\\path\\\\to\\\\folder」、「/path/to/folder」）です。",
+    "embeddingErrorDefault": "このエラーは、リポジトリを分析するために使用されるドキュメント埋め込みシステムに関連しています。モデル設定で異なる埋め込みプロバイダーを試してみてください。",
     "backToHome": "ホームに戻る",
     "exportWiki": "Wikiをエクスポート",
     "exportAsMarkdown": "Markdownとしてエクスポート",

src/messages/kr.json

@@ -103,6 +103,7 @@
       "home": "홈",
       "errorTitle": "오류",
       "errorMessageDefault": "저장소가 존재하며 공개 상태인지 확인해 주세요. 유효한 형식은 \"owner/repo\", \"https://github.com/owner/repo\", \"https://gitlab.com/owner/repo\", \"https://bitbucket.org/owner/repo\" 또는 로컬 폴더 경로 \"C:\\\\path\\\\to\\\\folder\" 혹은 \"/path/to/folder\" 입니다.",
+      "embeddingErrorDefault": "이 오류는 저장소를 분석하는 데 사용되는 문서 임베딩 시스템과 관련이 있습니다. 임베딩 모델 설정에서 다른 임베딩 제공자를 시도해 보세요. 문제가 지속되면 모델 설정에서 다른 임베딩 제공자를 변경해 보세요.",
       "backToHome": "홈으로 돌아가기",
       "exportWiki": "위키 내보내기",
       "exportAsMarkdown": "마크다운으로 내보내기",

src/messages/pt-br.json

@@ -112,6 +112,7 @@
     "home": "Início",
     "errorTitle": "Erro",
     "errorMessageDefault": "Verifique se o seu repositório existe e é público. Formatos válidos são \"proprietário/repo\", \"https://github.com/proprietário/repo\", \"https://gitlab.com/proprietário/repo\", \"https://bitbucket.org/proprietário/repo\", ou caminhos de pastas locais como \"C:\\\\caminho\\\\para\\\\pasta\" ou \"/caminho/para/pasta\".",
+    "embeddingErrorDefault": "Este erro está relacionado com o sistema de embebimento utilizado para analisar o seu repositório. Verifique a configuração do modelo de embebimento, as chaves de API e tente novamente. Se o problema persistir, considere mudar para um provedor de embebimento diferente na configuração do modelo.",
     "backToHome": "Voltar ao Início",
     "exportWiki": "Exportar Wiki",
     "exportAsMarkdown": "Exportar como Markdown",

src/messages/vi.json

@@ -103,6 +103,7 @@
     "home": "Trang chủ",
     "errorTitle": "Lỗi",
     "errorMessageDefault": "Vui lòng kiểm tra xem repository có tồn tại và công khai hay không. Các định dạng hợp lệ là \"owner/repo\", \"https://github.com/owner/repo\", \"https://gitlab.com/owner/repo\", \"https://bitbucket.org/owner/repo\", hoặc các đường dẫn thư mục cục bộ như \"C:\\\\path\\\\to\\\\folder\" hoặc \"/path/to/folder\".",
+    "embeddingErrorDefault": "Lỗi này liên quan đến hệ thống embedding được sử dụng để phân tích repository của bạn. Vui lòng kiểm tra cấu hình mô hình embedding, API keys và thử lại. Nếu vấn đề vẫn tiếp diễn, hãy xem xét chuyển sang nhà cung cấp embedding khác trong cấu hình mô hình.",
     "backToHome": "Quay lại trang chủ",
     "exportWiki": "Xuất Wiki",
     "exportAsMarkdown": "Xuất dưới dạng Markdown",

src/messages/zh-tw.json

@@ -101,6 +101,7 @@
     "home": "首頁",
     "errorTitle": "錯誤",
     "errorMessageDefault": "請檢查您的儲存庫是否存在且為公開儲存庫。有效格式為 \"owner/repo\"、\"https://github.com/owner/repo\"、\"https://gitlab.com/owner/repo\"、\"https://bitbucket.org/owner/repo\"，或本機資料夾路徑，如 \"C:\\\\path\\\\to\\\\folder\" 或 \"/path/to/folder\"。",
+    "embeddingErrorDefault": "這個錯誤與用於分析您的儲存庫的文件嵌入系統有關。請檢查您的嵌入模型配置、API 密鑰，並重試。如果問題持續存在，請考慮在模型設置中切換到不同的嵌入提供者。",
     "backToHome": "返回首頁",
     "exportWiki": "匯出 Wiki",
     "exportAsMarkdown": "匯出為 Markdown",