Support empty lines in BIO parser

nerval/parse.py

@@ -72,6 +72,9 @@ def parse_bio(lines: List[str]) -> dict:
     containing_tag = None
 
     for index, line in enumerate(lines):
+        if not line:
+            continue
+
         word, label = parse_line(index, line)
 
         # Preserve hyphens to avoid confusion with the hyphens added later during alignment

tests/conftest.py

@@ -10,6 +10,11 @@ def fake_annot_bio():
     return FIXTURES / "test_annot.bio"
 
 
+@pytest.fixture()
+def fake_annot_with_empty_lines_bio():
+    return FIXTURES / "test_annot_with_empty_lines.bio"
+
+
 @pytest.fixture()
 def fake_predict_bio():
     return FIXTURES / "test_predict.bio"

tests/fixtures/test_annot_with_empty_lines.bio

+Gérard B-PER
+de I-PER
+
+Nerval I-PER
+was O
+born O
+
+in O
+Paris B-LOC
+in O
+
+1808 B-DAT
+. O

tests/test_parse_bio.py

@@ -175,6 +175,7 @@ expected_parsed_end_of_file = {
         (pytest.lazy_fixture("fake_annot_bio"), expected_parsed_annot),
         (pytest.lazy_fixture("fake_predict_bio"), expected_parsed_predict),
         (pytest.lazy_fixture("empty_bio"), None),
+        (pytest.lazy_fixture("fake_annot_with_empty_lines_bio"), expected_parsed_annot),
         (pytest.lazy_fixture("bioeslu_bio"), expected_parsed_annot),
         (pytest.lazy_fixture("end_of_file_bio"), expected_parsed_end_of_file),
     ],