Added local Repo Support

sage/chat.py

@@ -19,7 +19,6 @@
 
 load_dotenv()
 
-
 def build_rag_chain(args):
     """Builds a RAG chain via LangChain."""
     llm = build_llm_via_langchain(args.llm_provider, args.llm_model)

sage/config.py

@@ -81,6 +81,12 @@ def add_repo_args(parser: ArgumentParser) -> Callable:
         default="repos",
         help="The local directory to store the repository",
     )
+    parser.add(
+        "--repo-mode",
+        default = "remote",
+        choices=["local", "remote"],
+        help="Define where is the repo present"
+    )
     return validate_repo_args
 
 
@@ -248,9 +254,12 @@ def validate_all(args):
 
 
 def validate_repo_args(args):
-    """Validates the configuration of the repository."""
-    if not re.match(r"^[^/]+/[^/]+$", args.repo_id):
-        raise ValueError("repo_id must be in the format 'owner/repo'")
+    """Validates the configuration of the repository.
+    For remote repositories, validates that repo_id is in 'owner/repo' format.
+    For local mode, accepts a single directory path.
+    """
+    if args.repo_mode != "local" and not re.match(r"^[^/]+/[^/]+$", args.repo_id):
+        raise ValueError("repo_id must be in the format 'owner/repo' for remote repositories")
 
 
 def _validate_openai_embedding_args(args):

sage/data_manager.py

@@ -96,6 +96,7 @@ def download(self) -> bool:
         """Clones the repository to the local directory, if it's not already cloned."""
         if os.path.exists(self.local_path):
             # The repository is already cloned.
+            logging.info("Repository already exists..")
             return True
 
         if not self.is_public and not self.access_token:
@@ -254,3 +255,5 @@ def from_args(args: Dict):
                 "For private repositories, please set the GITHUB_TOKEN variable in your environment."
             )
         return repo_manager
+
+

sage/embedder.py

@@ -420,4 +420,4 @@ def build_batch_embedder_from_flags(data_manager: DataManager, chunker: Chunker,
     elif args.embedding_provider == "gemini":
         return GeminiBatchEmbedder(data_manager, chunker, embedding_model=args.embedding_model)
     else:
-        raise ValueError(f"Unrecognized embedder type {args.embedding_provider}")
+        raise ValueError(f"Unrecognized embedder type {args.embedding_provider}")

sage/index.py

@@ -32,27 +32,30 @@ def main():
     ]
 
     args = parser.parse_args()
-
+    
     for validator in arg_validators:
         validator(args)
 
     if args.llm_retriever:
         logging.warning("The LLM retriever does not require indexing, so this script is a no-op.")
         return
-
+    
     # Additionally validate embedder and vector store compatibility.
     if args.embedding_provider == "openai" and args.vector_store_provider != "pinecone":
         parser.error("When using OpenAI embedder, the vector store type must be Pinecone.")
     if args.embedding_provider == "marqo" and args.vector_store_provider != "marqo":
         parser.error("When using the marqo embedder, the vector store type must also be marqo.")
-
+    if args.repo_mode == "local" and args.local_dir == "repos":
+        parser.error("You must not store the local repo inside the repos folder")
+
     ######################
     # Step 1: Embeddings #
     ######################
-
+    
     # Index the repository.
     repo_embedder = None
     if args.index_repo:
+        # Check the repo-mode
         logging.info("Cloning the repository...")
         repo_manager = GitHubRepoManager.from_args(args)
         logging.info("Embedding the repo...")