vortexa 0.1.3__tar.gz → 0.2.0__tar.gz

{vortexa-0.1.3 → vortexa-0.2.0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: vortexa
-Version: 0.1.3
+Version: 0.2.0
 Summary: Codebase indexing and semantic search engine
 Author-email: VortexAI <koulabhay25@gmail.com>
 License-Expression: Apache-2.0
@@ -28,12 +28,10 @@ Requires-Dist: pathspec>=0.12.0
 Requires-Dist: huggingface-hub>=0.20.0
 Requires-Dist: tokenizers>=0.19.0
 Requires-Dist: safetensors>=0.4.0
-Provides-Extra: full
-Requires-Dist: model2vec>=0.3.0; extra == "full"
-Requires-Dist: sentence-transformers>=2.2.0; extra == "full"
-Requires-Dist: tree-sitter-language-pack>=0.1.0; extra == "full"
-Provides-Extra: mcp
-Requires-Dist: fastmcp>=2.0.0; extra == "mcp"
+Requires-Dist: fastmcp>=2.0.0
+Requires-Dist: model2vec>=0.3.0
+Requires-Dist: sentence-transformers>=2.2.0
+Requires-Dist: tree-sitter-language-pack>=0.1.0
 Dynamic: license-file
 
 <div align="center">

{vortexa-0.1.3 → vortexa-0.2.0}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "vortexa"
-version = "0.1.3"
+version = "0.2.0"
 description = "Codebase indexing and semantic search engine"
 readme = "README.md"
 requires-python = ">=3.10"
@@ -34,17 +34,12 @@ dependencies = [
     "huggingface-hub>=0.20.0",
     "tokenizers>=0.19.0",
     "safetensors>=0.4.0",
-]
-
-[project.optional-dependencies]
-full = [
+    "fastmcp>=2.0.0",
     "model2vec>=0.3.0",
     "sentence-transformers>=2.2.0",
     "tree-sitter-language-pack>=0.1.0",
 ]
-mcp = [
-    "fastmcp>=2.0.0",
-]
+
 
 [project.scripts]
 vortexa = "vortexa.interfaces.cli:main"

{vortexa-0.1.3 → vortexa-0.2.0}/src/vortexa/core/chunking.py

@@ -31,7 +31,7 @@ class ChunkBoundary:
 def _get_parser(language: str):
     """Get a tree-sitter parser for the given language. Returns None if unavailable."""
     try:
-        from tree_sitter_language_pack import get_parser as _get_ts_parser  # ty: ignore[unresolved-import]
+        from tree_sitter_language_pack import get_parser as _get_ts_parser
 
         return _get_ts_parser(language)
     except Exception:

{vortexa-0.1.3 → vortexa-0.2.0}/src/vortexa/core/embedding.py

@@ -61,7 +61,7 @@ class Model2VecEmbedder:
         if self._model is None:
             with self._lock:
                 if self._model is None:  # Double-checked locking
-                    from model2vec import StaticModel  # ty: ignore[unresolved-import]
+                    from model2vec import StaticModel
                     logger.info("Loading embedding model: %s", self._model_id)
                     self._model = StaticModel.from_pretrained(self._model_id)
 
@@ -111,7 +111,7 @@ class SentenceTransformerEmbedder:
         if self._model is None:
             with self._lock:
                 if self._model is None:
-                    from sentence_transformers import SentenceTransformer  # ty: ignore[unresolved-import]
+                    from sentence_transformers import SentenceTransformer
                     logger.info("Loading sentence-transformers model: %s", self._model_name)
                     self._model = SentenceTransformer(self._model_name, device=self._device)
 

{vortexa-0.1.3 → vortexa-0.2.0}/src/vortexa/core/indexer.py

@@ -42,6 +42,19 @@ _MAX_FILE_BYTES = 1_000_000  # 1 MB
 # Hash of the chunking logic for memoization — bump when chunking changes
 _CHUNKING_LOGIC_VERSION = "2"
 
+_GLOBAL_INDEX_ROOT = Path.home() / ".vortexa"
+
+
+def _index_dir_for_root(root: Path) -> Path:
+    """Derive the global index directory for a given project root.
+
+    Stores under ``~/.vortexa/<hex_hash>/`` where *hex_hash* is the first
+    16 chars of the SHA-256 of the resolved absolute path.  This keeps every
+    indexed project isolated while avoiding path-length issues.
+    """
+    path_hash = hashlib.sha256(str(root).encode()).hexdigest()[:16]
+    return _GLOBAL_INDEX_ROOT / path_hash
+
 
 def _file_hash(path: Path) -> str:
     """Compute SHA256 hash of file contents."""
@@ -80,11 +93,11 @@ class CodebaseIndexer:
         :param root: Root directory of the codebase to index.
         :param model: Embedding model (Encoder or Embedder). Created from model_id if None.
         :param model_id: Model ID for auto-creating embedder (ignored if model provided).
-        :param index_dir: Directory for persistent storage. Defaults to .jarvis/index/.
+        :param index_dir: Directory for persistent storage. Defaults to ~/.vortexa/<hash>/.
         :param chunk_config: Chunking configuration.
         """
         self.root = Path(root).resolve()
-        self.index_dir = Path(index_dir) if index_dir else self.root / ".jarvis" / "index"
+        self.index_dir = Path(index_dir) if index_dir else _index_dir_for_root(self.root)
         self.chunk_config = chunk_config or ChunkConfig()
 
         # Resolve model: prefer explicit model, else create LF4Embedder

{vortexa-0.1.3 → vortexa-0.2.0}/src/vortexa/interfaces/cli.py

@@ -8,7 +8,7 @@ import logging
 import sys
 from pathlib import Path
 
-from vortexa.core.indexer import CodebaseIndexer
+from vortexa.core.indexer import CodebaseIndexer, _index_dir_for_root
 
 logger = logging.getLogger(__name__)
 
@@ -112,7 +112,7 @@ def _build_parser() -> argparse.ArgumentParser:
     parser.add_argument(
         "environment_details",
         nargs="?",
-        help="optional root path, JSON environment details, or Kilo environment details",
+        help="optional root path, JSON environment details, or text containing a root path to index and search (defaults to current directory)",
     )
     parser.add_argument(
         "--root",
@@ -138,12 +138,12 @@ def _build_parser() -> argparse.ArgumentParser:
     parser.add_argument(
         "--force",
         action="store_true",
-        help="force a full re-index before searching",
+        help="force a full re-index before searching (ignores existing index)",
     )
     parser.add_argument(
         "--no-index",
         action="store_true",
-        help="skip indexing and search the existing index only",
+        help="skip indexing entirely and search the existing index (errors if none)",
     )
     parser.add_argument(
         "--plain",
@@ -153,6 +153,12 @@ def _build_parser() -> argparse.ArgumentParser:
     return parser
 
 
+def _has_existing_index(root: Path) -> bool:
+    """Check if a persistent index already exists for this root."""
+    index_dir = _index_dir_for_root(root)
+    return (index_dir / "state.lmdb").exists()
+
+
 def run_query(args: argparse.Namespace, parser: argparse.ArgumentParser) -> int:
     if not args.query:
         raise SystemExit("query is required when using -q/--query")
@@ -167,15 +173,48 @@ def run_query(args: argparse.Namespace, parser: argparse.ArgumentParser) -> int:
         parser.error(str(exc))
     indexer = CodebaseIndexer(root=root)
 
+    has_index = _has_existing_index(root)
+
     if args.no_index:
+        if not has_index:
+            raise SystemExit(
+                f"No index found for {root}. Run without --no-index to build one."
+            )
         indexer._load_state()
-    else:
-        stats = indexer.index(force=args.force, include_text_files=args.include_text)
+        print(f"[vortexa] Loaded existing index ({len(indexer.chunks)} chunks)", file=sys.stderr)
+    elif args.force:
+        stats = indexer.index(force=True, include_text_files=args.include_text)
         print(
-            f"[vortexa] Ready: {stats.indexed_files} files, "
+            f"[vortexa] Re-indexed: {stats.indexed_files} files, "
             f"{stats.total_chunks} chunks in {stats.index_time_ms:.0f}ms",
             file=sys.stderr,
         )
+    else:
+        if not has_index:
+            stats = indexer.index(include_text_files=args.include_text)
+            print(
+                f"[vortexa] Indexed: {stats.indexed_files} files, "
+                f"{stats.total_chunks} chunks in {stats.index_time_ms:.0f}ms",
+                file=sys.stderr,
+            )
+        else:
+            indexer._load_state()
+            before = len(indexer.chunks)
+            stats = indexer.index(include_text_files=args.include_text)
+            after = len(indexer.chunks)
+            if after == before and stats.memo_misses == 0:
+                print(
+                    f"[vortexa] Index up to date ({after} chunks, no changes)",
+                    file=sys.stderr,
+                )
+            else:
+                added = after - before
+                label = f"+{added} chunks" if added else f"{stats.memo_misses} updated"
+                print(
+                    f"[vortexa] Updated: {stats.indexed_files} files, "
+                    f"{after} chunks ({label}) in {stats.index_time_ms:.0f}ms",
+                    file=sys.stderr,
+                )
 
     query = args.query
     results = indexer.search(query, top_k=args.top_k, alpha=args.alpha)

{vortexa-0.1.3 → vortexa-0.2.0}/src/vortexa/interfaces/mcp_server.py

@@ -14,7 +14,7 @@ from pathlib import Path
 
 os.environ.setdefault("HF_HUB_DISABLE_SYMLINKS_WARNING", "1")
 
-from fastmcp import FastMCP  # ty: ignore[unresolved-import]
+from fastmcp import FastMCP  # ty: ignore[unresolved-import, unused-ignore-comment]
 
 logger = logging.getLogger(__name__)
 

{vortexa-0.1.3 → vortexa-0.2.0}/src/vortexa/storage/walker.py

@@ -34,7 +34,6 @@ _DEFAULT_IGNORED_DIRS: frozenset[str] = frozenset(
         "dist/",
         "build/",
         ".eggs/",
-        ".jarvis/",
     }
 )
 

{vortexa-0.1.3 → vortexa-0.2.0}/src/vortexa.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: vortexa
-Version: 0.1.3
+Version: 0.2.0
 Summary: Codebase indexing and semantic search engine
 Author-email: VortexAI <koulabhay25@gmail.com>
 License-Expression: Apache-2.0
@@ -28,12 +28,10 @@ Requires-Dist: pathspec>=0.12.0
 Requires-Dist: huggingface-hub>=0.20.0
 Requires-Dist: tokenizers>=0.19.0
 Requires-Dist: safetensors>=0.4.0
-Provides-Extra: full
-Requires-Dist: model2vec>=0.3.0; extra == "full"
-Requires-Dist: sentence-transformers>=2.2.0; extra == "full"
-Requires-Dist: tree-sitter-language-pack>=0.1.0; extra == "full"
-Provides-Extra: mcp
-Requires-Dist: fastmcp>=2.0.0; extra == "mcp"
+Requires-Dist: fastmcp>=2.0.0
+Requires-Dist: model2vec>=0.3.0
+Requires-Dist: sentence-transformers>=2.2.0
+Requires-Dist: tree-sitter-language-pack>=0.1.0
 Dynamic: license-file
 
 <div align="center">

{vortexa-0.1.3 → vortexa-0.2.0}/src/vortexa.egg-info/requires.txt

@@ -5,11 +5,7 @@ pathspec>=0.12.0
 huggingface-hub>=0.20.0
 tokenizers>=0.19.0
 safetensors>=0.4.0
-
-[full]
+fastmcp>=2.0.0
 model2vec>=0.3.0
 sentence-transformers>=2.2.0
 tree-sitter-language-pack>=0.1.0
-
-[mcp]
-fastmcp>=2.0.0