fix: handling for empty tables in word docs and powerpoints (#982)

* fix table index error

* changelog and version

CHANGELOG.md

@@ -9,6 +9,8 @@
 
 ### Fixes
 
+* Handling for empty tables in Word Documents and PowerPoints.
+
 ## 0.8.4
 
 ### Enhancements

test_unstructured/partition/test_common.py

@@ -219,3 +219,14 @@ def test_convert_office_doc_captures_errors(monkeypatch, caplog):
     monkeypatch.setattr(subprocess, "Popen", MockPopenWithError)
     common.convert_office_doc("no-real.docx", "fake-directory", target_format="docx")
     assert "an error occurred" in caplog.text
+
+
+class MockDocxEmptyTable:
+    def __init__(self):
+        self.rows = []
+
+
+def test_convert_ms_office_table_to_text_works_with_empty_tables():
+    table = MockDocxEmptyTable()
+    assert common.convert_ms_office_table_to_text(table, as_html=True) == ""
+    assert common.convert_ms_office_table_to_text(table, as_html=False) == ""

unstructured/partition/common.py

@@ -318,6 +318,10 @@ def convert_ms_office_table_to_text(table: docxtable.Table, as_html: bool = True
     """
     fmt = "html" if as_html else "plain"
     rows = list(table.rows)
-    headers = [cell.text for cell in rows[0].cells]
-    data = [[cell.text for cell in row.cells] for row in rows[1:]]
-    return tabulate(data, headers=headers, tablefmt=fmt)
+    if len(rows) > 0:
+        headers = [cell.text for cell in rows[0].cells]
+        data = [[cell.text for cell in row.cells] for row in rows[1:]]
+        table_text = tabulate(data, headers=headers, tablefmt=fmt)
+    else:
+        table_text = ""
+    return table_text