pythongh-124969: Fix locale.nl_langinfo(locale.ALT_DIGITS)

Author: serhiy-storchaka

Doc/library/locale.rst

@@ -158,7 +158,8 @@ The :mod:`locale` module defines the following exception and functions:
 
 .. function:: nl_langinfo(option)
 
-   Return some locale-specific information as a string.  This function is not
+   Return some locale-specific information as a string (or a tuple for
+   ``ALT_DIGITS``).  This function is not
    available on all systems, and the set of possible options might also vary
    across platforms.  The possible argument values are numbers, for which
    symbolic constants are available in the locale module.
@@ -311,8 +312,7 @@ The :mod:`locale` module defines the following exception and functions:
 
    .. data:: ALT_DIGITS
 
-      Get a representation of up to 100 values used to represent the values
-      0 to 99.
+      Get a tuple of up to 100 strings used to represent the values 0 to 99.
 
 
 .. function:: getdefaultlocale([envvars])

Lib/test/test__locale.py

@@ -1,4 +1,4 @@
-from _locale import (setlocale, LC_ALL, LC_CTYPE, LC_NUMERIC, localeconv, Error)
+from _locale import (setlocale, LC_ALL, LC_CTYPE, LC_NUMERIC, LC_TIME, localeconv, Error)
 try:
     from _locale import (RADIXCHAR, THOUSEP, nl_langinfo)
 except ImportError:
@@ -74,6 +74,17 @@ def accept(loc):
     'ps_AF': ('\u066b', '\u066c'),
 }
 
+known_alt_digits = {
+    'C': (0, {}),
+    'en_US': (0, {}),
+    'fa_IR': (100, {0: '\u06f0\u06f0', 10: '\u06f1\u06f0', 99: '\u06f9\u06f9'}),
+    'ja_JP': (100, {0: '\u3007', 10: '\u5341', 99: '\u4e5d\u5341\u4e5d'}),
+    'lzh_TW': (32, {0: '\u3007', 10: '\u5341', 31: '\u5345\u4e00'}),
+    'my_MM': (100, {0: '\u1040\u1040', 10: '\u1041\u1040', 99: '\u1049\u1049'}),
+    'or_IN': (100, {0: '\u0b66', 10: '\u0b67\u0b66', 99: '\u0b6f\u0b6f'}),
+    'shn_MM': (100, {0: '\u1090\u1090', 10: '\u1091\u1090', 99: '\u1099\u1099'}),
+}
+
 if sys.platform == 'win32':
     # ps_AF doesn't work on Windows: see bpo-38324 (msg361830)
     del known_numerics['ps_AF']
@@ -176,6 +187,31 @@ def test_lc_numeric_basic(self):
         if not tested:
             self.skipTest('no suitable locales')
 
+    @unittest.skipUnless(nl_langinfo, "nl_langinfo is not available")
+    @unittest.skipUnless(hasattr(locale, 'ALT_DIGITS'), "requires locale.ALT_DIGITS")
+    @unittest.skipIf(
+        support.is_emscripten or support.is_wasi,
+        "musl libc issue on Emscripten, bpo-46390"
+    )
+    def test_alt_digits_nl_langinfo(self):
+        # Test nl_langinfo(ALT_DIGITS)
+        tested = False
+        for loc, (count, samples) in known_alt_digits.items():
+            try:
+                setlocale(LC_TIME, loc)
+                setlocale(LC_CTYPE, loc)
+            except Error:
+                continue
+            with self.subTest(locale=loc):
+                alt_digits = nl_langinfo(locale.ALT_DIGITS)
+                self.assertIsInstance(alt_digits, tuple)
+                self.assertEqual(len(alt_digits), count)
+                for i in samples:
+                    self.assertEqual(alt_digits[i], samples[i])
+                tested = True
+        if not tested:
+            self.skipTest('no suitable locales')
+
     def test_float_parsing(self):
         # Bug #1391872: Test whether float parsing is okay on European
         # locales.

Modules/_localemodule.c

@@ -608,6 +608,30 @@ _locale_nl_langinfo_impl(PyObject *module, int item)
                instead of an empty string for nl_langinfo(ERA).  */
             const char *result = nl_langinfo(item);
             result = result != NULL ? result : "";
+#ifdef ALT_DIGITS
+            if (item == ALT_DIGITS) {
+                /* The result is a sequence of up to 100 NUL-separated strings. */
+                const char *s = result;
+                int count = 0;
+                for (; count < 100 && *s; count++) {
+                    s += strlen(s) + 1;
+                }
+                PyObject *tuple = PyTuple_New(count);
+                if (tuple == NULL) {
+                    return NULL;
+                }
+                for (int i = 0; i < count; i++) {
+                    PyObject *unicode = PyUnicode_DecodeLocale(result, NULL);
+                    if (unicode == NULL) {
+                        Py_DECREF(tuple);
+                        return NULL;
+                    }
+                    PyTuple_SET_ITEM(tuple, i, unicode);
+                    result += strlen(result) + 1;
+                }
+                return tuple;
+            }
+#endif
             return PyUnicode_DecodeLocale(result, NULL);
         }
     PyErr_SetString(PyExc_ValueError, "unsupported langinfo constant");