Test languages exemplars canonical duplicates

googlefonts · Nov 1, 2022 · 036b804 · 036b804
1 parent a46f0ee
commit 036b804
Showing 1 changed file with 20 additions and 0 deletions.
diff --git a/tests/test_data_languages.py b/tests/test_data_languages.py
@@ -22,6 +22,26 @@
 LANGUAGES = LoadLanguages()
 
 
+@pytest.mark.parametrize("lang_code", LANGUAGES)
+@pytest.mark.parametrize(
+    "exemplar_name",
+    ["base", "auxiliary", "marks", "numerals", "punctuation", "index"]
+)
+def test_languages_exemplars_canonical_duplicates(lang_code, exemplar_name):
+    lang = LANGUAGES[lang_code]
+    exemplar = getattr(lang.exemplar_chars, exemplar_name).split()
+    normalized = defaultdict(set)
+
+    for g in exemplar:
+        if g[0] == "{" and g[-1] == "}":
+            g = g.lstrip("{").rstrip("}")
+        normalized[unicodedata.normalize("NFC", g)].add(g)
+
+    result = [(len(gs), gs) for n, gs in normalized.items()]
+    expected = [(1, {n}) for n, gs in normalized.items()]
+    assert result == expected
+
+
 @pytest.mark.parametrize("lang_code", LANGUAGES)
 @pytest.mark.parametrize(
     "exemplar_name",