directory loader improvements (#1162)

docs/modules/document_loaders/examples/directory_loader.ipynb

@@ -6,7 +6,7 @@
    "metadata": {},
    "source": [
     "# Directory Loader\n",
-    "This covers how to use the DirectoryLoader to load all documents in a directory. Under the hood, this uses the [UnstructuredLoader](./unstructured_file.ipynb)"
+    "This covers how to use the DirectoryLoader to load all documents in a directory. Under the hood, by default this uses the [UnstructuredLoader](./unstructured_file.ipynb)"
    ]
   },
   {
@@ -29,7 +29,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 7,
+   "execution_count": 2,
    "id": "891fe56f",
    "metadata": {},
    "outputs": [],
@@ -39,7 +39,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 8,
+   "execution_count": 3,
    "id": "addfe9cf",
    "metadata": {},
    "outputs": [],
@@ -49,7 +49,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 9,
+   "execution_count": 4,
    "id": "b042086d",
    "metadata": {},
    "outputs": [
@@ -59,7 +59,67 @@
        "1"
       ]
      },
-     "execution_count": 9,
+     "execution_count": 4,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "len(docs)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "c5652850",
+   "metadata": {},
+   "source": [
+    "## Change loader class\n",
+    "By default this uses the UnstructuredLoader class. However, you can change up the type of loader pretty easily."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "81c92da3",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from langchain.document_loaders import TextLoader"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "ab38ee36",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "loader = DirectoryLoader('../', glob=\"**/*.md\", loader_cls=TextLoader)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 7,
+   "id": "25c8740f",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "docs = loader.load()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 8,
+   "id": "38337763",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/plain": [
+       "1"
+      ]
+     },
+     "execution_count": 8,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -71,7 +131,7 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "id": "cbc8256b",
+   "id": "984c8429",
    "metadata": {},
    "outputs": [],
    "source": []

langchain/document_loaders/directory.py

@@ -1,19 +1,42 @@
 """Loading logic for loading documents from a directory."""
+import logging
 from pathlib import Path
-from typing import List
+from typing import List, Type, Union
 
 from langchain.docstore.document import Document
 from langchain.document_loaders.base import BaseLoader
+from langchain.document_loaders.text import TextLoader
 from langchain.document_loaders.unstructured import UnstructuredFileLoader
 
+FILE_LOADER_TYPE = Union[Type[UnstructuredFileLoader], Type[TextLoader]]
+logger = logging.getLogger(__file__)
+
+
+def _is_visible(p: Path) -> bool:
+    parts = p.parts
+    for _p in parts:
+        if _p.startswith("."):
+            return False
+    return True
+
 
 class DirectoryLoader(BaseLoader):
     """Loading logic for loading documents from a directory."""
 
-    def __init__(self, path: str, glob: str = "**/*"):
+    def __init__(
+        self,
+        path: str,
+        glob: str = "**/[!.]*",
+        silent_errors: bool = False,
+        load_hidden: bool = False,
+        loader_cls: FILE_LOADER_TYPE = UnstructuredFileLoader,
+    ):
         """Initialize with path to directory and how to glob over it."""
         self.path = path
         self.glob = glob
+        self.load_hidden = load_hidden
+        self.loader_cls = loader_cls
+        self.silent_errors = silent_errors
 
     def load(self) -> List[Document]:
         """Load documents."""
@@ -21,6 +44,13 @@ class DirectoryLoader(BaseLoader):
         docs = []
         for i in p.glob(self.glob):
             if i.is_file():
-                sub_docs = UnstructuredFileLoader(str(i)).load()
-                docs.extend(sub_docs)
+                if _is_visible(i.relative_to(p)) or self.load_hidden:
+                    try:
+                        sub_docs = self.loader_cls(str(i)).load()
+                        docs.extend(sub_docs)
+                    except Exception as e:
+                        if self.silent_errors:
+                            logger.warning(e)
+                        else:
+                            raise e
         return docs