add tests for more coverage

test_unstructured/metrics/test_element_type.py

@@ -40,6 +40,12 @@ def test_get_element_type_frequency(filename, frequency):
     assert elements_freq == frequency
 
 
+def test_get_element_type_frequency_zero_len():
+    elements = partition(filename=f"example-docs/blank.xlsx")
+    elements_freq = get_element_type_frequency(elements_to_json(elements))
+    assert len(elements_freq) == 0
+
+
 @pytest.mark.parametrize(
     ("filename", "expected_frequency", "percent_matched"),
     [
@@ -107,3 +113,11 @@ def test_calculate_element_type_percent_match(filename, expected_frequency, perc
         round(calculate_element_type_percent_match(elements_frequency, expected_frequency, 0.8), 2)
         == percent_matched[2]
     )
+
+
+def test_calculate_element_type_percent_match_zero_source_output():
+    with_frequency = {("Header", None): 1}
+    elements = partition(filename=f"example-docs/blank.xlsx")
+    no_frequency = get_element_type_frequency(elements_to_json(elements))
+    assert calculate_element_type_percent_match(with_frequency, no_frequency) == 0.0
+    assert calculate_element_type_percent_match(no_frequency, with_frequency) == 0.0

test_unstructured/metrics/test_text_extraction.py

@@ -216,3 +216,10 @@ def test_calculate_percent_missing_text(output_text, source_text, expected_perce
         text_extraction.calculate_percent_missing_text(output_text, source_text)
         == expected_percentage
     )
+
+
+def test_error_return_type():
+    output_elements = partition(filename=f"example-docs/fake-text.txt")
+    source_elements = partition(filename=f"example-docs/fake-text.txt")
+    with pytest.raises(ValueError):
+        text_extraction.calculate_edit_distance(output_elements, source_elements, "typo")

test_unstructured/partition/xlsx/test_xlsx.py

@@ -226,6 +226,12 @@ def test_partition_xlsx_subtables(filename="example-docs/vodafone.xlsx"):
     assert len(elements) == 6
 
 
+def test_partition_xlsx_not_find_subtable(filename="example-docs/vodafone.xlsx"):
+    elements = partition_xlsx(filename, find_subtable=False)
+    assert sum(isinstance(element, Table) for element in elements) == 1
+    assert len(elements) == 1
+
+
 def test_partition_xlsx_element_metadata_has_languages():
     filename = "example-docs/stanley-cups.xlsx"
     elements = partition_xlsx(filename=filename)