Fix: partition on empty or whitespace-only text files (#3675)

This is a fix for this
[bug](https://github.com/Unstructured-IO/unstructured/issues/3674), auto partition fails on text files which are empty or contain only whitespaces

Inference of .txt file type fails if the file has only whitespaces.

To Reproduce:

```
from tempfile import NamedTemporaryFile

from unstructured.partition.auto import partition

with NamedTemporaryFile(mode="w", suffix=".txt") as f:
    f.write("   \n")
    f.seek(0)
    elements = partition(filename=f.name)
```

CHANGELOG.md

@@ -1,4 +1,4 @@
-## 0.15.14-dev4
+## 0.15.14-dev5
 
 ### Enhancements
 
@@ -11,6 +11,7 @@
 * **Update Python SDK usage in `partition_via_api`.** Make a minor syntax change to ensure forward compatibility with the upcoming 0.26.0 Python SDK.
 * **Remove "unused" `date_from_file_object` parameter.** As part of simplifying partitioning parameter set, remove `date_from_file_object` parameter. A file object does not have a last-modified date attribute so can never give a useful value. When a file-object is used as the document source (such as in Unstructured API) the last-modified date must come from the `metadata_last_modified` argument.
 * **Fix occasional `KeyError` when mapping parent ids to hash ids.** Occasionally the input elements into `assign_and_map_hash_ids` can contain duplicated element instances, which lead to error when mapping parent id.
+* **Allow empty text files.** Fixes an issue where text files with only white space would fail to be partitioned.
 
 ## 0.15.13
 

example-docs/fake-text-all-whitespace.txt

@@ -0,0 +1,3 @@
+
+
+

test_unstructured/partition/test_auto.py

@@ -749,22 +749,30 @@ def test_auto_partition_tsv_from_filename():
 # ================================================================================================
 # TXT
 # ================================================================================================
-
-
-def test_auto_partition_text_from_filename():
-    file_path = example_doc_path("fake-text.txt")
-
-    elements = partition(filename=file_path, strategy=PartitionStrategy.HI_RES)
-
-    assert elements == [
+@pytest.mark.parametrize(
+    ("filename", "expected_elements"),
+    [
+        (
+            "fake-text.txt",
+            [
                 NarrativeText(text="This is a test document to use for unit tests."),
                 Address(text="Doylestown, PA 18901"),
                 Title(text="Important points:"),
                 ListItem(text="Hamburgers are delicious"),
                 ListItem(text="Dogs are the best"),
                 ListItem(text="I love fuzzy blankets"),
-    ]
-    assert all(e.metadata.filename == "fake-text.txt" for e in elements)
+            ],
+        ),
+        ("fake-text-all-whitespace.txt", []),
+    ],
+)
+def test_auto_partition_text_from_filename(filename: str, expected_elements: list[Element]):
+    file_path = example_doc_path(filename)
+
+    elements = partition(filename=file_path, strategy=PartitionStrategy.HI_RES)
+
+    assert elements == expected_elements
+    assert all(e.metadata.filename == filename for e in elements)
     assert all(e.metadata.file_directory == example_doc_path("") for e in elements)
 
 

unstructured/__version__.py

@@ -1 +1 @@
-__version__ = "0.15.14-dev4"  # pragma: no cover
+__version__ = "0.15.14-dev5"  # pragma: no cover

unstructured/file_utils/filetype.py

@@ -601,7 +601,7 @@ class _TextFileDifferentiator:
         text_head = self._ctx.text_head
 
         # -- an empty file is not JSON --
-        if not text_head:
+        if not text_head.lstrip():
             return False
 
         # -- has to be a list or object, no string, number, or bool --