Fixes after CR

Unstructured-IO · Nov 20, 2024 · 3ec8714 · 3ec8714
1 parent 85dc0fc
commit 3ec8714
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 3 deletions.
diff --git a/test_unstructured/chunking/test_html_output.py b/test_unstructured/chunking/test_html_output.py
@@ -21,7 +21,7 @@ def test_combining_html_metadata_when_multiple_elements_in_composite_element(chu
         '<input class="FormFieldValue" id="5" value="Example value" />'
         "</form>"
     )
-    combined_metadata = metadata_1 + metadata_2 + metadata_3
+    combined_metadata = " ".join([metadata_1, metadata_2, metadata_3])
 
     elements = [
         Title(text="Header", metadata=ElementMetadata(text_as_html=metadata_1)),
@@ -71,7 +71,7 @@ def test_combining_html_metadata_with_nested_relationship_between_elements(chunk
     assert chunks[0].text == "First"
     assert chunks[1].text == "Second"
 
-    assert chunks[0].metadata.text_as_html == metadata_1 + metadata_2
+    assert chunks[0].metadata.text_as_html == metadata_1 + " " + metadata_2
     assert chunks[1].metadata.text_as_html == metadata_3
 
 

diff --git a/unstructured/chunking/base.py b/unstructured/chunking/base.py
@@ -775,7 +775,7 @@ def iter_kwarg_pairs() -> Iterator[tuple[str, Any]]:
                     ordered_unique_keys = {key: None for val_list in values for key in val_list}
                     yield field_name, list(ordered_unique_keys.keys())
                 elif strategy is CS.STRING_CONCATENATE:
-                    yield field_name, "".join(values)
+                    yield field_name, " ".join(val.strip() for val in values)
                 elif strategy is CS.DROP:
                     continue
                 else:  # pragma: no cover