hecvec 0.4.4__tar.gz → 0.4.6__tar.gz

{hecvec-0.4.4 → hecvec-0.4.6}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: hecvec
-Version: 0.4.4
+Version: 0.4.6
 Summary: List directories (safe root), filter .txt/.md files, read as text, chunk, embed, and push to Chroma.
 License-Expression: MIT
 Keywords: chunking,document-pipeline,listdir,text-files

{hecvec-0.4.4 → hecvec-0.4.6}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "hecvec"
-version = "0.4.4"
+version = "0.4.6"
 description = "List directories (safe root), filter .txt/.md files, read as text, chunk, embed, and push to Chroma."
 readme = "README.md"
 requires-python = ">=3.9,<3.14"

{hecvec-0.4.4 → hecvec-0.4.6}/src/hecvec/__init__.py

@@ -35,4 +35,4 @@ __all__ = [
     "__version__",
 ]
 
-__version__ = "0.4.4"
+__version__ = "0.4.6"

{hecvec-0.4.4 → hecvec-0.4.6}/src/hecvec/pipeline.py

@@ -57,6 +57,8 @@ class Slicer:
         collection_name: str = "hecvec",
         chroma_host: str = "localhost",
         chroma_port: int = 8000,
+        chroma_mode: ChromaMode = "auto",
+        chroma_persist_path: str | Path | None = None,
         embedding_model: str = "text-embedding-3-small",
         chunk_size: int = 200,
         chunk_overlap: int = 0,
@@ -215,7 +217,7 @@ class Slicer:
 
         # 5/5 Push to Chroma
         stage_start = perf_counter()
-        logger.info("[5/5] Writing to Chroma | host=%s | port=%s | collection=%s", chroma_host, chroma_port, collection_name)
+        logger.info("[5/5] Chroma | host=%s | port=%s | collection=%s", chroma_host, chroma_port, collection_name)
         client, chroma_mode_used = get_client(
             host=chroma_host,
             port=chroma_port,
@@ -223,11 +225,11 @@ class Slicer:
             mode=chroma_mode,
         )
         if chroma_mode_used == "server":
-            logger.info("[5/5] Using Chroma server at %s:%s", chroma_host, chroma_port)
+            logger.info("[5/5] Chroma: connected to server at %s:%s (data can persist)", chroma_host, chroma_port)
         elif chroma_mode_used == "persistent":
-            logger.info("[5/5] Using Chroma persistent storage at %s", chroma_persist_path)
+            logger.info("[5/5] Chroma: using persistent storage at %s (data persists on disk)", chroma_persist_path)
         else:
-            logger.info("[5/5] Using Chroma in-memory (ephemeral) client")
+            logger.info("[5/5] Chroma: not connected to server; using in-memory (ephemeral). Data is lost when this process exits.")
         logger.info("[5/5] Adding %d document chunk(s) to collection...", len(documents))
         add_result = add_documents(client, collection_name, ids, embeddings, documents)
         if add_result["collection_existed"]: