Merge branch 'main' into test-out-of-script

diff --cc tests/test_data_languages.py
index 7dcd6f4dccc547b10a4c3938a1ece07535b1e0ca,17870d442eae2a9c639590863dca187118535761..315275d953ed277f3e63fffff3ab2a0c2abba350
--- 1/tests/test_data_languages.py
--- 2/tests/test_data_languages.py
+++ b/tests/test_data_languages.py
@@@ -195,53 -182,15 +201,67 @@@ def test_exemplars_are_in_script(lang_c
      )
  
  
 +@pytest.mark.parametrize("lang_code", LANGUAGES.keys())
 +def test_sample_texts_are_in_script(lang_code):
 +    if lang_code in [
 +        "mak_Maka",
 +        "orv_Cyrl",
 +        "cu_Cyrl",
 +        "ff_Adlm",
 +        "idu_Latn",
 +        "ban_Bali",
 +    ]:
 +        pytest.xfail("These languages have known issues with their sample text")
 +        return
 +    lang = LANGUAGES[lang_code]
 +    script_name = SCRIPTS[lang.script].name
 +    script_name = CLDR_SCRIPT_TO_UCD_SCRIPT.get(script_name, script_name)
 +    if not lang.sample_text.ListFields():
 +        pytest.skip("No sample text for language " + lang_code)
 +        return
 +    if lang.id in SKIP_EXEMPLARS:
 +        pytest.skip(SKIP_EXEMPLARS[lang.id])
 +        return
 +    out_of_script = defaultdict(set)
 +    for field in SampleText.fields:
 +        if field.name == "note":
 +            continue
 +        samples = getattr(lang.sample_text, field.name)
 +        chars = set(samples)
 +        for char in chars:
 +            char_script = (
 +                youseedee.ucd_data(ord(char)).get("Script", "").replace("_", " ")
 +            )
 +            if char_script == "Common" or char_script == "Inherited":
 +                continue
 +            if char_script != script_name:
 +                extensions = (
 +                    youseedee.ucd_data(ord(char))
 +                    .get("Script_Extensions", "")
 +                    .split(" ")
 +                )
 +                if any(ext == lang.script for ext in extensions):
 +                    continue
 +                out_of_script[char_script].add(char)
 +                break
 +    msg = []
 +    for script, chars in out_of_script.items():
 +        msg.append(f"'{''.join(chars)}' ({script} != {script_name})")
 +    assert not out_of_script, (
 +        f"{lang_code} sample text contained out-of-script characters"
 +        f": {', '.join(msg)}"
 +    )
++
++
+ def test_exemplar_parser():
+     bases = "a A ā Ā {a̍} {A̍} {kl}"
+     parsed_bases = parse(bases)
+     assert parsed_bases == {
+         "a",
+         "A",
+         "ā",
+         "Ā",
+         "k",
+         "l",
+         "̍",
+     }