fix: raises on file-like object with .name not a valid path (#2614)

**Summary**
Fixes: #2308

**Additional context**
Through a somewhat deep call-chain, partitioning a file-like object
(e.g. io.BytesIO) having its `.name` attribute set to a path not
pointing to an actual file on the local filesystem would raise
`FileNotFoundError` when the last-modified date was being computed for
the document.

This scenario is a legitimate partitioning call, where `file.name` is
used downstream to describe the source of, for example, a bytes payload
downloaded from the network.

**Fix**
- explicitly check for the existence of a file at the given path before
accessing it to get its modified date. Return `None` (already a
legitimate return value) when no such file exists.
- Generally clean up the implementations.
- Add unit tests that exercise all cases.

---------

Co-authored-by: John <43506685+Coniferish@users.noreply.github.com>

CHANGELOG.md

@@ -1,4 +1,4 @@
-## 0.12.6-dev7
+## 0.12.6-dev8
 
 ### Enhancements
 
@@ -10,6 +10,7 @@
 
 ### Fixes
 
+* **Partitioning raises on file-like object with `.name` not a local file path.** When partitioning a file using the `file=` argument, and `file` is a file-like object (e.g. io.BytesIO) having a `.name` attribute, and the value of `file.name` is not a valid path to a file present on the local filesystem, `FileNotFoundError` is raised. This prevents use of the `file.name` attribute for downstream purposes to, for example, describe the source of a document retrieved from a network location via HTTP.
 * **Fix SharePoint dates with inconsistent formatting** Adds logic to conditionally support dates returned by office365 that may vary in date formatting or may be a datetime rather than a string.
 * **Include warnings** about the potential risk of installing a version of `pandoc` which does not support RTF files + instructions that will help resolve that issue.
 * **Incorporate the `install-pandoc` Makefile recipe** into relevant stages of CI workflow, ensuring it is a version that supports RTF input files.

test_unstructured/partition/test_common.py

@@ -1,3 +1,7 @@
+import datetime as dt
+import io
+import os
+import pathlib
 from dataclasses import dataclass
 from unittest import mock
 
@@ -540,3 +544,64 @@ def test_ocr_data_to_elements(
             points=layout_el.bbox.coordinates,
             system=coordinate_system,
         )
+
+
+class Describe_get_last_modified_date:
+
+    def it_gets_the_modified_time_of_a_file_identified_by_a_path(self, tmp_path: pathlib.Path):
+        modified_timestamp = dt.datetime(
+            year=2024, month=3, day=5, hour=17, minute=43, second=40
+        ).timestamp()
+        file_path = tmp_path / "some_file.txt"
+        file_path.write_text("abcdefg")
+        os.utime(file_path, (modified_timestamp, modified_timestamp))
+
+        last_modified_date = common.get_last_modified_date(str(file_path))
+
+        assert last_modified_date == "2024-03-05T17:43:40"
+
+    def but_it_returns_None_when_there_is_no_file_at_that_path(self, tmp_path: pathlib.Path):
+        file_path = tmp_path / "some_file_that_does_not_exist.txt"
+
+        last_modified_date = common.get_last_modified_date(str(file_path))
+
+        assert last_modified_date is None
+
+
+class Describe_get_last_modified_date_from_file:
+
+    def it_gets_the_modified_time_of_a_file_like_object_corresponding_to_a_filesystem_file(
+        self, tmp_path: pathlib.Path
+    ):
+        modified_timestamp = dt.datetime(
+            year=2024, month=3, day=5, hour=20, minute=48, second=26
+        ).timestamp()
+        file_path = tmp_path / "some_file_2.txt"
+        file_path.write_text("abcdefg")
+        os.utime(file_path, (modified_timestamp, modified_timestamp))
+
+        with open(file_path, "rb") as f:
+            last_modified_date = common.get_last_modified_date_from_file(f)
+
+        assert last_modified_date == "2024-03-05T20:48:26"
+
+    def but_it_returns_None_when_the_argument_is_a_bytes_object(self):
+        assert common.get_last_modified_date_from_file(b"abcdefg") is None
+
+    def and_it_returns_None_when_the_file_like_object_has_no_name_attribute(self):
+        file = io.BytesIO(b"abcdefg")
+        assert hasattr(file, "name") is False
+
+        last_modified_date = common.get_last_modified_date_from_file(file)
+
+        assert last_modified_date is None
+
+    def and_it_returns_None_when_the_file_like_object_name_is_not_a_path_to_a_file(
+        self, tmp_path: pathlib.Path
+    ):
+        file = io.BytesIO(b"abcdefg")
+        file.name = str(tmp_path / "a_file_that_isn't_here.txt")
+
+        last_modified_date = common.get_last_modified_date_from_file(file)
+
+        assert last_modified_date is None

test_unstructured/partition/xlsx/test_xlsx.py

@@ -4,6 +4,7 @@
 
 from __future__ import annotations
 
+import io
 import sys
 import tempfile
 from typing import cast
@@ -163,6 +164,23 @@ def test_partition_xlsx_from_file():
     assert elements[1].metadata.filename is None
 
 
+def test_partition_xlsx_from_file_like_object_with_name():
+    with open("example-docs/stanley-cups.xlsx", "rb") as f:
+        file = io.BytesIO(f.read())
+    file.name = "stanley-cups-downloaded-from-network.xlsx"
+
+    elements = partition_xlsx(file=file, include_header=False)
+
+    assert sum(isinstance(element, Table) for element in elements) == 2
+    assert len(elements) == 4
+    assert clean_extra_whitespace(elements[0].text) == EXPECTED_TITLE
+    assert clean_extra_whitespace(elements[1].text) == EXPECTED_TEXT_XLSX
+    assert elements[1].metadata.text_as_html == EXPECTED_TABLE_XLSX
+    assert elements[1].metadata.page_number == 1
+    assert elements[1].metadata.filetype == EXPECTED_FILETYPE
+    assert elements[1].metadata.page_name == EXCEPTED_PAGE_NAME
+
+
 def test_partition_xlsx_from_file_with_metadata_filename():
     with open("example-docs/stanley-cups.xlsx", "rb") as f:
         elements = partition_xlsx(file=f, metadata_filename="test", include_header=False)

unstructured/__version__.py

@@ -1 +1 @@
-__version__ = "0.12.6-dev7"  # pragma: no cover
+__version__ = "0.12.6-dev8"  # pragma: no cover

unstructured/partition/common.py

@@ -75,21 +75,27 @@ def get_last_modified_date(filename: str) -> Optional[str]:
     Otherwise returns date and time in ISO 8601 string format (YYYY-MM-DDTHH:MM:SS) like
     "2024-03-05T17:02:53".
     """
+    if not os.path.isfile(filename):
+        return None
+
     modify_date = datetime.fromtimestamp(os.path.getmtime(filename))
     return modify_date.strftime("%Y-%m-%dT%H:%M:%S%z")
 
 
 def get_last_modified_date_from_file(file: IO[bytes] | bytes) -> Optional[str]:
     """Modified timestamp of `file` if it corresponds to a file on the local filesystem."""
-    filename = None
-    if hasattr(file, "name"):
-        filename = file.name
+    # -- a file-like object will have a name attribute if created by `open()` or if a name is
+    # -- assigned to it for metadata purposes. Use "" as default because the empty string is never
+    # -- a path to an actual file.
+    filename = str(getattr(file, "name", ""))
 
-    if not filename:
+    # -- there's no guarantee the path corresponds to an actual file on the filesystem. In
+    # -- particular, a user can set the `.name` attribute of an e.g. `io.BytesIO` object to
+    # -- populate the `.metadata.filename` fields for a payload perhaps downloaded via HTTP.
+    if not os.path.isfile(filename):
         return None
 
-    modify_date = get_last_modified_date(filename)
-    return modify_date
+    return get_last_modified_date(filename)
 
 
 def normalize_layout_element(
@@ -145,7 +151,7 @@ def normalize_layout_element(
             )
 
     elif element_type in TYPE_TO_TEXT_ELEMENT_MAP:
-        assert isinstance(element_type, str)
+        assert isinstance(element_type, str)  # Added to resolve type-error
         _element_class = TYPE_TO_TEXT_ELEMENT_MAP[element_type]
         _element_class = _element_class(
             text=text,