Pauses Latin testing in lieu of #514 (#519)

* [mlt] Updates Maltese phonelist.

Due either to bugs or changes in the upstream data, I noticed there was
a very high rate of filtration on Maltese. It seems that [u] was not
included, nor was one of the affricates.

There are still some filtration for "archaic" pronunciations of
[ɣ] for <għ>, which is WAI.

* Changelog

* Adds Python 3.12 support

* project classifers
* tests on CircleCI

* Revert "Adds Python 3.12 support"

This reverts commit e72bc3d.

* Pauses Latin testing in lieu of #514.

* Fixes typo in test_split

* More explicit comments.

* changelog

* Reruns black

* black

CHANGELOG.md

@@ -128,6 +128,7 @@ Unreleased
 
 #### Added
 
+-  Temporarily disables Latin testing in lieu of #514. (\#519)
 -  Fixed dialect selectors for languages other than Latin. (\#511)
 -  Moved `wikipron/` directory under `src/` and adjusted package finding. (\#508)
 -  Added documentation about selecting transcription level. (\#502)

tests/test_data/test_split.py

@@ -165,7 +165,8 @@
         [("ᤀᤠᤀᤡᤴᤋᤠᤴᤍᤡᤰ", True), ("ژלرنال", False), ("wikipron", False)],
     ),
     SmokeTestScript(
-        "Bopomofo", [("ㄅㄆㄇㄈ", True), ("ژלرنال", False), ("wikipron", False)]
+        "Bopomofo",
+        [("ㄅㄆㄇㄈ", True), ("ژלرنال", False), ("wikipron", False)],
     ),
     SmokeTestScript(
         "Georgian",

tests/test_wikipron/test_scrape.py

@@ -35,7 +35,9 @@
     SmokeTestLanguage("khm", "Khmer", {}),
     SmokeTestLanguage("shn", "Shan", {}),
     SmokeTestLanguage("tha", "Thai", {}),
-    SmokeTestLanguage("lat", "Latin", {}),
+    # TODO(#514): Latin extractor is broken; disabling. Uncomment the
+    # following line to re-enable.
+    # SmokeTestLanguage("lat", "Latin", {}),
     # Japanese data is mostly narrow transcription.
     SmokeTestLanguage("jpn", "Japanese", {"narrow": True}),
     SmokeTestLanguage("cmn", "Chinese", {"skip_spaces_pron": False}),
@@ -77,6 +79,9 @@ def test_special_languages_covered_by_smoke_test():
     """All languages handled by wikipron.extract must have a smoke test."""
     special_languages = {lang for lang in EXTRACTION_FUNCTIONS.keys()}
     smoke_test_languages = {lang.wik_name for lang in _SMOKE_TEST_LANGUAGES}
+    # TODO(#514): Latin extractor is broken; disabling its check here. Remove
+    # the following line to re-enable.
+    smoke_test_languages.add("Latin")
     assert special_languages.issubset(smoke_test_languages), (
         "These languages must also be included in the smoke test: "
         f"{special_languages - smoke_test_languages}"