Fix edge cases with empty reference

.github/workflows/test-pip-install.yaml

@@ -51,13 +51,13 @@ jobs:
       - name: Install kaldialign
         shell: bash
         run: |
-          pip3 install --verbose kaldialign
+          pip3 install --verbose 'kaldialign[test]'
 
       - name: Run test
         shell: bash
         run: |
           cd tests
           python3 -c "import kaldialign; print(kaldialign.__file__)"
           python3 -c "import kaldialign; print(kaldialign.__version__)"
+          pytest .
 
-          python3 ./test_align.py

.github/workflows/test.yaml

@@ -43,4 +43,4 @@ jobs:
       - name: Test
         shell: bash
         run: |
-          python3 ./tests/test_align.py
+          pytest -vv ./tests/test_align.py

CMakeLists.txt

@@ -14,7 +14,7 @@ cmake_minimum_required(VERSION 3.8 FATAL_ERROR)
 project(kaldialign CXX)
 
 # Please remember to also change line 3 of ./scripts/conda/kaldialign/meta.yaml
-set(KALDIALIGN_VERSION "0.9")
+set(KALDIALIGN_VERSION "0.9.1")
 
 if(NOT CMAKE_BUILD_TYPE)
   set(CMAKE_BUILD_TYPE Release)

kaldialign/__init__.py

@@ -38,7 +38,13 @@ def edit_distance(
 
     ans = _kaldialign.edit_distance(refi, hypi, sclite_mode)
     ans["ref_len"] = len(refi)
-    ans["err_rate"] = ans["total"] / len(refi)
+    try:
+        ans["err_rate"] = ans["total"] / len(refi)
+    except ZeroDivisionError:
+        if ans["total"] == 0:
+            ans["err_rate"] = 0.0
+        else:
+            ans["err_rate"] = float("inf")
     return ans
 
 

scripts/conda/kaldialign/meta.yaml

@@ -1,6 +1,6 @@
 package:
   name: kaldialign
-  version: "0.9"
+  version: "0.9.1"
 
 source:
   path: "{{ environ.get('KALDIALIGN_ROOT_DIR') }}"

tests/test_align.py

@@ -86,6 +86,34 @@ def test_edit_distance():
     }
 
 
+def test_edit_distance_zero_len_ref_zero_err():
+    a = []
+    b = []
+    results = edit_distance(a, b)
+    assert results == {
+        "ins": 0,
+        "del": 0,
+        "sub": 0,
+        "total": 0,
+        "ref_len": 0,
+        "err_rate": 0,
+    }
+
+
+def test_edit_distance_zero_len_ref_with_err():
+    a = []
+    b = ["a"]
+    results = edit_distance(a, b)
+    assert results == {
+        "ins": 1,
+        "del": 0,
+        "sub": 0,
+        "total": 1,
+        "ref_len": 0,
+        "err_rate": float("inf"),
+    }
+
+
 def test_edit_distance_sclite():
     a = ["a", "b"]
     b = ["b", "c"]
@@ -156,10 +184,3 @@ def test_bootstrap_wer_ci_2system():
 
     assert ans["p_s2_improv_over_s1"] == 1.0
 
-
-if __name__ == "__main__":
-    test_align()
-    test_edit_distance()
-    test_edit_distance_sclite()
-    test_bootstrap_wer_ci_1system()
-    test_bootstrap_wer_ci_2system()