From f980144e9c670a18119bf8127e184e82fa988f8b Mon Sep 17 00:00:00 2001
From: LIU Yuwei <22045841+Marsman1996@users.noreply.github.com>
Date: Mon, 13 Jan 2025 22:26:00 +0800
Subject: [PATCH] community: add init for unstructured file loader (#29101)

## Description
Add `__init__` for unstructured loader of
epub/image/markdown/pdf/ppt/word to restrict the input type to `str` or
`Path`.
In the
[signature](https://python.langchain.com/api_reference/community/document_loaders/langchain_community.document_loaders.markdown.UnstructuredMarkdownLoader.html)
these unstructured loaders receive `file_path: str | List[str] | Path |
List[Path]`, but actually they only receive `str` or `Path`.

## Issue
None

## Dependencies
No changes.
---
 .../document_loaders/chm.py                   | 20 +++++++++-
 .../document_loaders/epub.py                  | 29 ++++++++++----
 .../document_loaders/excel.py                 |  1 +
 .../document_loaders/image.py                 | 20 +++++++++-
 .../document_loaders/markdown.py              | 38 ++++++++++++-------
 .../document_loaders/odt.py                   |  1 +
 .../document_loaders/org_mode.py              |  1 +
 .../document_loaders/pdf.py                   | 17 +++++++++
 .../document_loaders/powerpoint.py            | 37 ++++++++++++------
 .../document_loaders/rst.py                   |  1 +
 .../document_loaders/rtf.py                   | 11 ++----
 .../document_loaders/tsv.py                   |  1 +
 .../document_loaders/word_document.py         | 36 ++++++++++++------
 13 files changed, 157 insertions(+), 56 deletions(-)

diff --git a/libs/community/langchain_community/document_loaders/chm.py b/libs/community/langchain_community/document_loaders/chm.py
index 207d31bd4f8..42ef6457bb8 100644
--- a/libs/community/langchain_community/document_loaders/chm.py
+++ b/libs/community/langchain_community/document_loaders/chm.py
@@ -1,4 +1,5 @@
-from typing import TYPE_CHECKING, Dict, List, Union
+from pathlib import Path
+from typing import TYPE_CHECKING, Any, Dict, List, Union
 
 from langchain_community.document_loaders.unstructured import UnstructuredFileLoader
 
@@ -24,6 +25,23 @@ class UnstructuredCHMLoader(UnstructuredFileLoader):
     http://www.jedrea.com/chmlib/
     """
 
+    def __init__(
+        self,
+        file_path: Union[str, Path],
+        mode: str = "single",
+        **unstructured_kwargs: Any,
+    ):
+        """
+
+        Args:
+            file_path: The path to the CHM file to load.
+            mode: The mode to use when loading the file. Can be one of "single",
+                "multi", or "all". Default is "single".
+            **unstructured_kwargs: Any kwargs to pass to the unstructured.
+        """
+        file_path = str(file_path)
+        super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
+
     def _get_elements(self) -> List:
         from unstructured.partition.html import partition_html
 
diff --git a/libs/community/langchain_community/document_loaders/epub.py b/libs/community/langchain_community/document_loaders/epub.py
index 6a316069127..d72fe0661e5 100644
--- a/libs/community/langchain_community/document_loaders/epub.py
+++ b/libs/community/langchain_community/document_loaders/epub.py
@@ -1,8 +1,9 @@
-from typing import List
+from pathlib import Path
+from typing import Any, List, Union
 
 from langchain_community.document_loaders.unstructured import (
     UnstructuredFileLoader,
-    satisfies_min_unstructured_version,
+    validate_unstructured_version,
 )
 
 
@@ -30,13 +31,25 @@ class UnstructuredEPubLoader(UnstructuredFileLoader):
     https://unstructured-io.github.io/unstructured/bricks.html#partition-epub
     """
 
+    def __init__(
+        self,
+        file_path: Union[str, Path],
+        mode: str = "single",
+        **unstructured_kwargs: Any,
+    ):
+        """
+
+        Args:
+            file_path: The path to the EPub file to load.
+            mode: The mode to use when loading the file. Can be one of "single",
+                "multi", or "all". Default is "single".
+            **unstructured_kwargs: Any kwargs to pass to the unstructured.
+        """
+        file_path = str(file_path)
+        validate_unstructured_version("0.5.4")
+        super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
+
     def _get_elements(self) -> List:
-        min_unstructured_version = "0.5.4"
-        if not satisfies_min_unstructured_version(min_unstructured_version):
-            raise ValueError(
-                "Partitioning epub files is only supported in "
-                f"unstructured>={min_unstructured_version}."
-            )
         from unstructured.partition.epub import partition_epub
 
         return partition_epub(filename=self.file_path, **self.unstructured_kwargs)  # type: ignore[arg-type]
diff --git a/libs/community/langchain_community/document_loaders/excel.py b/libs/community/langchain_community/document_loaders/excel.py
index ace4eeee646..aa28d891100 100644
--- a/libs/community/langchain_community/document_loaders/excel.py
+++ b/libs/community/langchain_community/document_loaders/excel.py
@@ -42,6 +42,7 @@ class UnstructuredExcelLoader(UnstructuredFileLoader):
               for more info. Optional. Defaults to "single".
             **unstructured_kwargs: Keyword arguments to pass to unstructured.
         """
+        file_path = str(file_path)
         validate_unstructured_version(min_unstructured_version="0.6.7")
         super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
 
diff --git a/libs/community/langchain_community/document_loaders/image.py b/libs/community/langchain_community/document_loaders/image.py
index c574226c753..63a89e71ccb 100644
--- a/libs/community/langchain_community/document_loaders/image.py
+++ b/libs/community/langchain_community/document_loaders/image.py
@@ -1,4 +1,5 @@
-from typing import List
+from pathlib import Path
+from typing import Any, List, Union
 
 from langchain_community.document_loaders.unstructured import UnstructuredFileLoader
 
@@ -27,6 +28,23 @@ class UnstructuredImageLoader(UnstructuredFileLoader):
     https://unstructured-io.github.io/unstructured/bricks.html#partition-image
     """
 
+    def __init__(
+        self,
+        file_path: Union[str, Path],
+        mode: str = "single",
+        **unstructured_kwargs: Any,
+    ):
+        """
+
+        Args:
+            file_path: The path to the Image file to load.
+            mode: The mode to use when loading the file. Can be one of "single",
+                "multi", or "all". Default is "single".
+            **unstructured_kwargs: Any kwargs to pass to the unstructured.
+        """
+        file_path = str(file_path)
+        super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
+
     def _get_elements(self) -> List:
         from unstructured.partition.image import partition_image
 
diff --git a/libs/community/langchain_community/document_loaders/markdown.py b/libs/community/langchain_community/document_loaders/markdown.py
index f204b0bdc03..3c3196c2cb4 100644
--- a/libs/community/langchain_community/document_loaders/markdown.py
+++ b/libs/community/langchain_community/document_loaders/markdown.py
@@ -1,6 +1,10 @@
-from typing import List
+from pathlib import Path
+from typing import Any, List, Union
 
-from langchain_community.document_loaders.unstructured import UnstructuredFileLoader
+from langchain_community.document_loaders.unstructured import (
+    UnstructuredFileLoader,
+    validate_unstructured_version,
+)
 
 
 class UnstructuredMarkdownLoader(UnstructuredFileLoader):
@@ -68,19 +72,25 @@ class UnstructuredMarkdownLoader(UnstructuredFileLoader):
     https://unstructured-io.github.io/unstructured/core/partition.html#partition-md
     """  # noqa: E501
 
+    def __init__(
+        self,
+        file_path: Union[str, Path],
+        mode: str = "single",
+        **unstructured_kwargs: Any,
+    ):
+        """
+
+        Args:
+            file_path: The path to the Markdown file to load.
+            mode: The mode to use when loading the file. Can be one of "single",
+                "multi", or "all". Default is "single".
+            **unstructured_kwargs: Any kwargs to pass to the unstructured.
+        """
+        file_path = str(file_path)
+        validate_unstructured_version("0.4.16")
+        super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
+
     def _get_elements(self) -> List:
-        from unstructured.__version__ import __version__ as __unstructured_version__
         from unstructured.partition.md import partition_md
 
-        # NOTE(MthwRobinson) - enables the loader to work when you're using pre-release
-        # versions of unstructured like 0.4.17-dev1
-        _unstructured_version = __unstructured_version__.split("-")[0]
-        unstructured_version = tuple([int(x) for x in _unstructured_version.split(".")])
-
-        if unstructured_version < (0, 4, 16):
-            raise ValueError(
-                f"You are on unstructured version {__unstructured_version__}. "
-                "Partitioning markdown files is only supported in unstructured>=0.4.16."
-            )
-
         return partition_md(filename=self.file_path, **self.unstructured_kwargs)  # type: ignore[arg-type]
diff --git a/libs/community/langchain_community/document_loaders/odt.py b/libs/community/langchain_community/document_loaders/odt.py
index a803aeafcbe..7d1236ff881 100644
--- a/libs/community/langchain_community/document_loaders/odt.py
+++ b/libs/community/langchain_community/document_loaders/odt.py
@@ -45,6 +45,7 @@ class UnstructuredODTLoader(UnstructuredFileLoader):
                 "multi", or "all". Default is "single".
             **unstructured_kwargs: Any kwargs to pass to the unstructured.
         """
+        file_path = str(file_path)
         validate_unstructured_version(min_unstructured_version="0.6.3")
         super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
 
diff --git a/libs/community/langchain_community/document_loaders/org_mode.py b/libs/community/langchain_community/document_loaders/org_mode.py
index 106bf58da48..b8ba1828433 100644
--- a/libs/community/langchain_community/document_loaders/org_mode.py
+++ b/libs/community/langchain_community/document_loaders/org_mode.py
@@ -45,6 +45,7 @@ class UnstructuredOrgModeLoader(UnstructuredFileLoader):
             **unstructured_kwargs: Any additional keyword arguments to pass
                 to the unstructured.
         """
+        file_path = str(file_path)
         validate_unstructured_version(min_unstructured_version="0.7.9")
         super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
 
diff --git a/libs/community/langchain_community/document_loaders/pdf.py b/libs/community/langchain_community/document_loaders/pdf.py
index c8ee848a733..25b3e72a3d8 100644
--- a/libs/community/langchain_community/document_loaders/pdf.py
+++ b/libs/community/langchain_community/document_loaders/pdf.py
@@ -68,6 +68,23 @@ class UnstructuredPDFLoader(UnstructuredFileLoader):
     https://unstructured-io.github.io/unstructured/bricks.html#partition-pdf
     """
 
+    def __init__(
+        self,
+        file_path: Union[str, Path],
+        mode: str = "single",
+        **unstructured_kwargs: Any,
+    ):
+        """
+
+        Args:
+            file_path: The path to the PDF file to load.
+            mode: The mode to use when loading the file. Can be one of "single",
+                "multi", or "all". Default is "single".
+            **unstructured_kwargs: Any kwargs to pass to the unstructured.
+        """
+        file_path = str(file_path)
+        super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
+
     def _get_elements(self) -> list:
         from unstructured.partition.pdf import partition_pdf
 
diff --git a/libs/community/langchain_community/document_loaders/powerpoint.py b/libs/community/langchain_community/document_loaders/powerpoint.py
index ed360e94f90..0fecbe4b678 100644
--- a/libs/community/langchain_community/document_loaders/powerpoint.py
+++ b/libs/community/langchain_community/document_loaders/powerpoint.py
@@ -1,7 +1,11 @@
 import os
-from typing import List
+from pathlib import Path
+from typing import Any, List, Union
 
-from langchain_community.document_loaders.unstructured import UnstructuredFileLoader
+from langchain_community.document_loaders.unstructured import (
+    UnstructuredFileLoader,
+    validate_unstructured_version,
+)
 
 
 class UnstructuredPowerPointLoader(UnstructuredFileLoader):
@@ -29,13 +33,26 @@ class UnstructuredPowerPointLoader(UnstructuredFileLoader):
     https://unstructured-io.github.io/unstructured/bricks.html#partition-pptx
     """
 
+    def __init__(
+        self,
+        file_path: Union[str, Path],
+        mode: str = "single",
+        **unstructured_kwargs: Any,
+    ):
+        """
+
+        Args:
+            file_path: The path to the PowerPoint file to load.
+            mode: The mode to use when loading the file. Can be one of "single",
+                "multi", or "all". Default is "single".
+            **unstructured_kwargs: Any kwargs to pass to the unstructured.
+        """
+        file_path = str(file_path)
+        super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
+
     def _get_elements(self) -> List:
-        from unstructured.__version__ import __version__ as __unstructured_version__
         from unstructured.file_utils.filetype import FileType, detect_filetype
 
-        unstructured_version = tuple(
-            [int(x) for x in __unstructured_version__.split(".")]
-        )
         # NOTE(MthwRobinson) - magic will raise an import error if the libmagic
         # system dependency isn't installed. If it's not installed, we'll just
         # check the file extension
@@ -47,12 +64,8 @@ class UnstructuredPowerPointLoader(UnstructuredFileLoader):
             _, extension = os.path.splitext(str(self.file_path))
             is_ppt = extension == ".ppt"
 
-        if is_ppt and unstructured_version < (0, 4, 11):
-            raise ValueError(
-                f"You are on unstructured version {__unstructured_version__}. "
-                "Partitioning .ppt files is only supported in unstructured>=0.4.11. "
-                "Please upgrade the unstructured package and try again."
-            )
+        if is_ppt:
+            validate_unstructured_version("0.4.11")
 
         if is_ppt:
             from unstructured.partition.ppt import partition_ppt
diff --git a/libs/community/langchain_community/document_loaders/rst.py b/libs/community/langchain_community/document_loaders/rst.py
index 310f1bb7c60..77e4fc38b91 100644
--- a/libs/community/langchain_community/document_loaders/rst.py
+++ b/libs/community/langchain_community/document_loaders/rst.py
@@ -49,6 +49,7 @@ class UnstructuredRSTLoader(UnstructuredFileLoader):
             **unstructured_kwargs: Additional keyword arguments to pass
                 to unstructured.
         """
+        file_path = str(file_path)
         validate_unstructured_version(min_unstructured_version="0.7.5")
         super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
 
diff --git a/libs/community/langchain_community/document_loaders/rtf.py b/libs/community/langchain_community/document_loaders/rtf.py
index 63128aa8372..bbecd7f455d 100644
--- a/libs/community/langchain_community/document_loaders/rtf.py
+++ b/libs/community/langchain_community/document_loaders/rtf.py
@@ -5,7 +5,7 @@ from typing import Any, List, Union
 
 from langchain_community.document_loaders.unstructured import (
     UnstructuredFileLoader,
-    satisfies_min_unstructured_version,
+    validate_unstructured_version,
 )
 
 
@@ -49,13 +49,8 @@ class UnstructuredRTFLoader(UnstructuredFileLoader):
             **unstructured_kwargs: Additional keyword arguments to pass
                 to unstructured.
         """
-        min_unstructured_version = "0.5.12"
-        if not satisfies_min_unstructured_version(min_unstructured_version):
-            raise ValueError(
-                "Partitioning rtf files is only supported in "
-                f"unstructured>={min_unstructured_version}."
-            )
-
+        file_path = str(file_path)
+        validate_unstructured_version("0.5.12")
         super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
 
     def _get_elements(self) -> List:
diff --git a/libs/community/langchain_community/document_loaders/tsv.py b/libs/community/langchain_community/document_loaders/tsv.py
index f6d4a085ce7..a9455fd06f4 100644
--- a/libs/community/langchain_community/document_loaders/tsv.py
+++ b/libs/community/langchain_community/document_loaders/tsv.py
@@ -32,6 +32,7 @@ class UnstructuredTSVLoader(UnstructuredFileLoader):
         mode: str = "single",
         **unstructured_kwargs: Any,
     ):
+        file_path = str(file_path)
         validate_unstructured_version(min_unstructured_version="0.7.6")
         super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
 
diff --git a/libs/community/langchain_community/document_loaders/word_document.py b/libs/community/langchain_community/document_loaders/word_document.py
index bd2c4ae8f57..eac96eb5faa 100644
--- a/libs/community/langchain_community/document_loaders/word_document.py
+++ b/libs/community/langchain_community/document_loaders/word_document.py
@@ -4,14 +4,17 @@ import os
 import tempfile
 from abc import ABC
 from pathlib import Path
-from typing import List, Union
+from typing import Any, List, Union
 from urllib.parse import urlparse
 
 import requests
 from langchain_core.documents import Document
 
 from langchain_community.document_loaders.base import BaseLoader
-from langchain_community.document_loaders.unstructured import UnstructuredFileLoader
+from langchain_community.document_loaders.unstructured import (
+    UnstructuredFileLoader,
+    validate_unstructured_version,
+)
 
 
 class Docx2txtLoader(BaseLoader, ABC):
@@ -92,13 +95,26 @@ class UnstructuredWordDocumentLoader(UnstructuredFileLoader):
     https://unstructured-io.github.io/unstructured/bricks.html#partition-docx
     """
 
+    def __init__(
+        self,
+        file_path: Union[str, Path],
+        mode: str = "single",
+        **unstructured_kwargs: Any,
+    ):
+        """
+
+        Args:
+            file_path: The path to the Word file to load.
+            mode: The mode to use when loading the file. Can be one of "single",
+                "multi", or "all". Default is "single".
+            **unstructured_kwargs: Any kwargs to pass to the unstructured.
+        """
+        file_path = str(file_path)
+        super().__init__(file_path=file_path, mode=mode, **unstructured_kwargs)
+
     def _get_elements(self) -> List:
-        from unstructured.__version__ import __version__ as __unstructured_version__
         from unstructured.file_utils.filetype import FileType, detect_filetype
 
-        unstructured_version = tuple(
-            [int(x) for x in __unstructured_version__.split(".")]
-        )
         # NOTE(MthwRobinson) - magic will raise an import error if the libmagic
         # system dependency isn't installed. If it's not installed, we'll just
         # check the file extension
@@ -110,12 +126,8 @@ class UnstructuredWordDocumentLoader(UnstructuredFileLoader):
             _, extension = os.path.splitext(str(self.file_path))
             is_doc = extension == ".doc"
 
-        if is_doc and unstructured_version < (0, 4, 11):
-            raise ValueError(
-                f"You are on unstructured version {__unstructured_version__}. "
-                "Partitioning .doc files is only supported in unstructured>=0.4.11. "
-                "Please upgrade the unstructured package and try again."
-            )
+        if is_doc:
+            validate_unstructured_version("0.4.11")
 
         if is_doc:
             from unstructured.partition.doc import partition_doc