Update Python from 3.11.8 to 3.12.2

16 files changed, 86 insertions, 76 deletions

diff --git a/contrib/tools/python3/src/Objects/stringlib/asciilib.h b/contrib/tools/python3/src/Objects/stringlib/asciilib.h
index eebe888e41..b3016bfbbb 100644
--- a/contrib/tools/python3/src/Objects/stringlib/asciilib.h
+++ b/contrib/tools/python3/src/Objects/stringlib/asciilib.h
@@ -21,6 +21,7 @@
 #define STRINGLIB_CHECK          PyUnicode_Check
 #define STRINGLIB_CHECK_EXACT    PyUnicode_CheckExact
 #define STRINGLIB_MUTABLE 0
+#define STRINGLIB_FAST_MEMCHR    memchr
 
 #define STRINGLIB_TOSTR          PyObject_Str
 #define STRINGLIB_TOASCII        PyObject_ASCII
diff --git a/contrib/tools/python3/src/Objects/stringlib/clinic/transmogrify.h.h b/contrib/tools/python3/src/Objects/stringlib/clinic/transmogrify.h.h
index b88517bd36..49388cf043 100644
--- a/contrib/tools/python3/src/Objects/stringlib/clinic/transmogrify.h.h
+++ b/contrib/tools/python3/src/Objects/stringlib/clinic/transmogrify.h.h
@@ -2,6 +2,12 @@
 preserve
 [clinic start generated code]*/
 
+#if defined(Py_BUILD_CORE) && !defined(Py_BUILD_CORE_MODULE)
+#  include "pycore_gc.h"            // PyGC_Head
+#  include "pycore_runtime.h"       // _Py_ID()
+#endif
+
+
 PyDoc_STRVAR(stringlib_expandtabs__doc__,
 "expandtabs($self, /, tabsize=8)\n"
 "--\n"
@@ -20,8 +26,31 @@ static PyObject *
 stringlib_expandtabs(PyObject *self, PyObject *const *args, Py_ssize_t nargs, PyObject *kwnames)
 {
     PyObject *return_value = NULL;
+    #if defined(Py_BUILD_CORE) && !defined(Py_BUILD_CORE_MODULE)
+
+    #define NUM_KEYWORDS 1
+    static struct {
+        PyGC_Head _this_is_not_used;
+        PyObject_VAR_HEAD
+        PyObject *ob_item[NUM_KEYWORDS];
+    } _kwtuple = {
+        .ob_base = PyVarObject_HEAD_INIT(&PyTuple_Type, NUM_KEYWORDS)
+        .ob_item = { &_Py_ID(tabsize), },
+    };
+    #undef NUM_KEYWORDS
+    #define KWTUPLE (&_kwtuple.ob_base.ob_base)
+
+    #else  // !Py_BUILD_CORE
+    #  define KWTUPLE NULL
+    #endif  // !Py_BUILD_CORE
+
     static const char * const _keywords[] = {"tabsize", NULL};
-    static _PyArg_Parser _parser = {NULL, _keywords, "expandtabs", 0};
+    static _PyArg_Parser _parser = {
+        .keywords = _keywords,
+        .fname = "expandtabs",
+        .kwtuple = KWTUPLE,
+    };
+    #undef KWTUPLE
     PyObject *argsbuf[1];
     Py_ssize_t noptargs = nargs + (kwnames ? PyTuple_GET_SIZE(kwnames) : 0) - 0;
     int tabsize = 8;
@@ -249,4 +278,4 @@ stringlib_zfill(PyObject *self, PyObject *arg)
 exit:
     return return_value;
 }
-/*[clinic end generated code: output=46d058103bffedf7 input=a9049054013a1b77]*/
+/*[clinic end generated code: output=d44a269805f6739e input=a9049054013a1b77]*/
diff --git a/contrib/tools/python3/src/Objects/stringlib/count.h b/contrib/tools/python3/src/Objects/stringlib/count.h
index f48500bf56..e20edcd104 100644
--- a/contrib/tools/python3/src/Objects/stringlib/count.h
+++ b/contrib/tools/python3/src/Objects/stringlib/count.h
@@ -4,6 +4,11 @@
 #error must include "stringlib/fastsearch.h" before including this module
 #endif
 
+// gh-97982: Implementing asciilib_count() is not worth it, FASTSEARCH() does
+// not specialize the code for ASCII strings. Use ucs1lib_count() for ASCII and
+// UCS1 strings: it's the same than asciilib_count().
+#if !STRINGLIB_IS_UNICODE || STRINGLIB_MAX_CHAR > 0x7Fu
+
 Py_LOCAL_INLINE(Py_ssize_t)
 STRINGLIB(count)(const STRINGLIB_CHAR* str, Py_ssize_t str_len,
                 const STRINGLIB_CHAR* sub, Py_ssize_t sub_len,
@@ -24,4 +29,4 @@ STRINGLIB(count)(const STRINGLIB_CHAR* str, Py_ssize_t str_len,
     return count;
 }
 
-
+#endif
diff --git a/contrib/tools/python3/src/Objects/stringlib/eq.h b/contrib/tools/python3/src/Objects/stringlib/eq.h
index 9c1058b86c..2eac4baf5c 100644
--- a/contrib/tools/python3/src/Objects/stringlib/eq.h
+++ b/contrib/tools/python3/src/Objects/stringlib/eq.h
@@ -4,16 +4,8 @@
  * unicode_eq() is called when the hash of two unicode objects is equal.
  */
 Py_LOCAL_INLINE(int)
-unicode_eq(PyObject *aa, PyObject *bb)
+unicode_eq(PyObject *a, PyObject *b)
 {
-    assert(PyUnicode_Check(aa));
-    assert(PyUnicode_Check(bb));
-    assert(PyUnicode_IS_READY(aa));
-    assert(PyUnicode_IS_READY(bb));
-
-    PyUnicodeObject *a = (PyUnicodeObject *)aa;
-    PyUnicodeObject *b = (PyUnicodeObject *)bb;
-
     if (PyUnicode_GET_LENGTH(a) != PyUnicode_GET_LENGTH(b))
         return 0;
     if (PyUnicode_GET_LENGTH(a) == 0)
diff --git a/contrib/tools/python3/src/Objects/stringlib/fastsearch.h b/contrib/tools/python3/src/Objects/stringlib/fastsearch.h
index 7403d8a3f7..257b7bd678 100644
--- a/contrib/tools/python3/src/Objects/stringlib/fastsearch.h
+++ b/contrib/tools/python3/src/Objects/stringlib/fastsearch.h
@@ -18,7 +18,8 @@
    algorithm, which has worst-case O(n) runtime and best-case O(n/k).
    Also compute a table of shifts to achieve O(n/k) in more cases,
    and often (data dependent) deduce larger shifts than pure C&P can
-   deduce. */
+   deduce. See stringlib_find_two_way_notes.txt in this folder for a
+   detailed explanation. */
 
 #define FAST_COUNT 0
 #define FAST_SEARCH 1
@@ -39,7 +40,7 @@
 #define STRINGLIB_BLOOM(mask, ch)     \
     ((mask &  (1UL << ((ch) & (STRINGLIB_BLOOM_WIDTH -1)))))
 
-#if STRINGLIB_SIZEOF_CHAR == 1
+#ifdef STRINGLIB_FAST_MEMCHR
 #  define MEMCHR_CUT_OFF 15
 #else
 #  define MEMCHR_CUT_OFF 40
@@ -53,8 +54,8 @@ STRINGLIB(find_char)(const STRINGLIB_CHAR* s, Py_ssize_t n, STRINGLIB_CHAR ch)
     p = s;
     e = s + n;
     if (n > MEMCHR_CUT_OFF) {
-#if STRINGLIB_SIZEOF_CHAR == 1
-        p = memchr(s, ch, n);
+#ifdef STRINGLIB_FAST_MEMCHR
+        p = STRINGLIB_FAST_MEMCHR(s, ch, n);
         if (p != NULL)
             return (p - s);
         return -1;
@@ -102,16 +103,26 @@ STRINGLIB(find_char)(const STRINGLIB_CHAR* s, Py_ssize_t n, STRINGLIB_CHAR ch)
     return -1;
 }
 
+#undef MEMCHR_CUT_OFF
+
+#if STRINGLIB_SIZEOF_CHAR == 1
+#  define MEMRCHR_CUT_OFF 15
+#else
+#  define MEMRCHR_CUT_OFF 40
+#endif
+
+
 Py_LOCAL_INLINE(Py_ssize_t)
 STRINGLIB(rfind_char)(const STRINGLIB_CHAR* s, Py_ssize_t n, STRINGLIB_CHAR ch)
 {
     const STRINGLIB_CHAR *p;
 #ifdef HAVE_MEMRCHR
-    /* memrchr() is a GNU extension, available since glibc 2.1.91.
-       it doesn't seem as optimized as memchr(), but is still quite
-       faster than our hand-written loop below */
+    /* memrchr() is a GNU extension, available since glibc 2.1.91.  it
+       doesn't seem as optimized as memchr(), but is still quite
+       faster than our hand-written loop below. There is no wmemrchr
+       for 4-byte chars. */
 
-    if (n > MEMCHR_CUT_OFF) {
+    if (n > MEMRCHR_CUT_OFF) {
 #if STRINGLIB_SIZEOF_CHAR == 1
         p = memrchr(s, ch, n);
         if (p != NULL)
@@ -139,11 +150,11 @@ STRINGLIB(rfind_char)(const STRINGLIB_CHAR* s, Py_ssize_t n, STRINGLIB_CHAR ch)
                 if (*p == ch)
                     return n;
                 /* False positive */
-                if (n1 - n > MEMCHR_CUT_OFF)
+                if (n1 - n > MEMRCHR_CUT_OFF)
                     continue;
-                if (n <= MEMCHR_CUT_OFF)
+                if (n <= MEMRCHR_CUT_OFF)
                     break;
-                s1 = p - MEMCHR_CUT_OFF;
+                s1 = p - MEMRCHR_CUT_OFF;
                 while (p > s1) {
                     p--;
                     if (*p == ch)
@@ -151,7 +162,7 @@ STRINGLIB(rfind_char)(const STRINGLIB_CHAR* s, Py_ssize_t n, STRINGLIB_CHAR ch)
                 }
                 n = p - s;
             }
-            while (n > MEMCHR_CUT_OFF);
+            while (n > MEMRCHR_CUT_OFF);
         }
 #endif
     }
@@ -165,7 +176,7 @@ STRINGLIB(rfind_char)(const STRINGLIB_CHAR* s, Py_ssize_t n, STRINGLIB_CHAR ch)
     return -1;
 }
 
-#undef MEMCHR_CUT_OFF
+#undef MEMRCHR_CUT_OFF
 
 /* Change to a 1 to see logging comments walk through the algorithm. */
 #if 0 && STRINGLIB_SIZEOF_CHAR == 1
@@ -388,7 +399,7 @@ STRINGLIB(_two_way)(const STRINGLIB_CHAR *haystack, Py_ssize_t len_haystack,
                 if (window_last >= haystack_end) {
                     return -1;
                 }
-                LOG("Horspool skip");
+                LOG("Horspool skip\n");
             }
           no_shift:
             window = window_last - len_needle + 1;
@@ -447,7 +458,7 @@ STRINGLIB(_two_way)(const STRINGLIB_CHAR *haystack, Py_ssize_t len_haystack,
                 if (window_last >= haystack_end) {
                     return -1;
                 }
-                LOG("Horspool skip");
+                LOG("Horspool skip\n");
             }
             window = window_last - len_needle + 1;
             assert((window[len_needle - 1] & TABLE_MASK) ==
diff --git a/contrib/tools/python3/src/Objects/stringlib/join.h b/contrib/tools/python3/src/Objects/stringlib/join.h
index bb011f7db7..de6bd83ffe 100644
--- a/contrib/tools/python3/src/Objects/stringlib/join.h
+++ b/contrib/tools/python3/src/Objects/stringlib/join.h
@@ -63,8 +63,7 @@ STRINGLIB(bytes_join)(PyObject *sep, PyObject *iterable)
         item = PySequence_Fast_GET_ITEM(seq, i);
         if (PyBytes_CheckExact(item)) {
             /* Fast path. */
-            Py_INCREF(item);
-            buffers[i].obj = item;
+            buffers[i].obj = Py_NewRef(item);
             buffers[i].buf = PyBytes_AS_STRING(item);
             buffers[i].len = PyBytes_GET_SIZE(item);
         }
diff --git a/contrib/tools/python3/src/Objects/stringlib/localeutil.h b/contrib/tools/python3/src/Objects/stringlib/localeutil.h
index bd16e0a172..d77715ec0d 100644
--- a/contrib/tools/python3/src/Objects/stringlib/localeutil.h
+++ b/contrib/tools/python3/src/Objects/stringlib/localeutil.h
@@ -75,7 +75,7 @@ InsertThousandsGrouping_fill(_PyUnicodeWriter *writer, Py_ssize_t *buffer_pos,
 
     if (n_zeros) {
         *buffer_pos -= n_zeros;
-        enum PyUnicode_Kind kind = PyUnicode_KIND(writer->buffer);
+        int kind = PyUnicode_KIND(writer->buffer);
         void *data = PyUnicode_DATA(writer->buffer);
         unicode_fill(kind, data, '0', *buffer_pos, n_zeros);
     }
diff --git a/contrib/tools/python3/src/Objects/stringlib/replace.h b/contrib/tools/python3/src/Objects/stringlib/replace.h
index ef318ed6dd..123c9f850f 100644
--- a/contrib/tools/python3/src/Objects/stringlib/replace.h
+++ b/contrib/tools/python3/src/Objects/stringlib/replace.h
@@ -29,9 +29,9 @@ STRINGLIB(replace_1char_inplace)(STRINGLIB_CHAR* s, STRINGLIB_CHAR* end,
                 if (!--attempts) {
                     /* if u1 was not found for attempts iterations,
                        use FASTSEARCH() or memchr() */
-#if STRINGLIB_SIZEOF_CHAR == 1
+#ifdef STRINGLIB_FAST_MEMCHR
                     s++;
-                    s = memchr(s, u1, end - s);
+                    s = STRINGLIB_FAST_MEMCHR(s, u1, end - s);
                     if (s == NULL)
                         return;
 #else
diff --git a/contrib/tools/python3/src/Objects/stringlib/stringdefs.h b/contrib/tools/python3/src/Objects/stringlib/stringdefs.h
index 88641b25d4..484b98b729 100644
--- a/contrib/tools/python3/src/Objects/stringlib/stringdefs.h
+++ b/contrib/tools/python3/src/Objects/stringlib/stringdefs.h
@@ -24,4 +24,5 @@
 #define STRINGLIB_CHECK_EXACT    PyBytes_CheckExact
 #define STRINGLIB_TOSTR          PyObject_Str
 #define STRINGLIB_TOASCII        PyObject_Repr
+#define STRINGLIB_FAST_MEMCHR    memchr
 #endif /* !STRINGLIB_STRINGDEFS_H */
diff --git a/contrib/tools/python3/src/Objects/stringlib/transmogrify.h b/contrib/tools/python3/src/Objects/stringlib/transmogrify.h
index e1165ea38e..71099bb586 100644
--- a/contrib/tools/python3/src/Objects/stringlib/transmogrify.h
+++ b/contrib/tools/python3/src/Objects/stringlib/transmogrify.h
@@ -17,8 +17,7 @@ return_self(PyObject *self)
 {
 #if !STRINGLIB_MUTABLE
     if (STRINGLIB_CHECK_EXACT(self)) {
-        Py_INCREF(self);
-        return self;
+        return Py_NewRef(self);
     }
 #endif
     return STRINGLIB_NEW(STRINGLIB_STR(self), STRINGLIB_LEN(self));
diff --git a/contrib/tools/python3/src/Objects/stringlib/ucs1lib.h b/contrib/tools/python3/src/Objects/stringlib/ucs1lib.h
index 026ab11f1f..1b9b65ecba 100644
--- a/contrib/tools/python3/src/Objects/stringlib/ucs1lib.h
+++ b/contrib/tools/python3/src/Objects/stringlib/ucs1lib.h
@@ -20,6 +20,7 @@
 #define STRINGLIB_NEW            _PyUnicode_FromUCS1
 #define STRINGLIB_CHECK          PyUnicode_Check
 #define STRINGLIB_CHECK_EXACT    PyUnicode_CheckExact
+#define STRINGLIB_FAST_MEMCHR    memchr
 #define STRINGLIB_MUTABLE 0
 
 #define STRINGLIB_TOSTR          PyObject_Str
diff --git a/contrib/tools/python3/src/Objects/stringlib/ucs2lib.h b/contrib/tools/python3/src/Objects/stringlib/ucs2lib.h
index 75f11bc290..4b49bbb31d 100644
--- a/contrib/tools/python3/src/Objects/stringlib/ucs2lib.h
+++ b/contrib/tools/python3/src/Objects/stringlib/ucs2lib.h
@@ -21,6 +21,10 @@
 #define STRINGLIB_CHECK          PyUnicode_Check
 #define STRINGLIB_CHECK_EXACT    PyUnicode_CheckExact
 #define STRINGLIB_MUTABLE 0
+#if SIZEOF_WCHAR_T == 2
+#define STRINGLIB_FAST_MEMCHR(s, c, n)              \
+    (Py_UCS2 *)wmemchr((const wchar_t *)(s), c, n)
+#endif
 
 #define STRINGLIB_TOSTR          PyObject_Str
 #define STRINGLIB_TOASCII        PyObject_ASCII
diff --git a/contrib/tools/python3/src/Objects/stringlib/ucs4lib.h b/contrib/tools/python3/src/Objects/stringlib/ucs4lib.h
index 57344f235b..def4ca5d17 100644
--- a/contrib/tools/python3/src/Objects/stringlib/ucs4lib.h
+++ b/contrib/tools/python3/src/Objects/stringlib/ucs4lib.h
@@ -21,6 +21,10 @@
 #define STRINGLIB_CHECK          PyUnicode_Check
 #define STRINGLIB_CHECK_EXACT    PyUnicode_CheckExact
 #define STRINGLIB_MUTABLE 0
+#if SIZEOF_WCHAR_T == 4
+#define STRINGLIB_FAST_MEMCHR(s, c, n)              \
+    (Py_UCS4 *)wmemchr((const wchar_t *)(s), c, n)
+#endif
 
 #define STRINGLIB_TOSTR          PyObject_Str
 #define STRINGLIB_TOASCII        PyObject_ASCII
diff --git a/contrib/tools/python3/src/Objects/stringlib/undef.h b/contrib/tools/python3/src/Objects/stringlib/undef.h
index bf32298505..cc873a2ec4 100644
--- a/contrib/tools/python3/src/Objects/stringlib/undef.h
+++ b/contrib/tools/python3/src/Objects/stringlib/undef.h
@@ -8,3 +8,4 @@
 #undef STRINGLIB_NEW
 #undef STRINGLIB_IS_UNICODE
 #undef STRINGLIB_MUTABLE
+#undef STRINGLIB_FAST_MEMCHR
diff --git a/contrib/tools/python3/src/Objects/stringlib/unicode_format.h b/contrib/tools/python3/src/Objects/stringlib/unicode_format.h
index a4eea7b919..ccd7c77c0a 100644
--- a/contrib/tools/python3/src/Objects/stringlib/unicode_format.h
+++ b/contrib/tools/python3/src/Objects/stringlib/unicode_format.h
@@ -473,8 +473,7 @@ get_field_object(SubString *input, PyObject *args, PyObject *kwargs,
             goto error;
 
         /* assign to obj */
-        Py_DECREF(obj);
-        obj = tmp;
+        Py_SETREF(obj, tmp);
     }
     /* end of iterator, this is the non-error case */
     if (ok == 1)
@@ -825,8 +824,7 @@ output_markup(SubString *field_name, SubString *format_spec,
             goto done;
 
         /* do the assignment, transferring ownership: fieldobj = tmp */
-        Py_DECREF(fieldobj);
-        fieldobj = tmp;
+        Py_SETREF(fieldobj, tmp);
         tmp = NULL;
     }
 
@@ -1042,8 +1040,7 @@ formatteriter_next(formatteriterobject *it)
            otherwise create a one length string with the conversion
            character */
         if (conversion == '\0') {
-            conversion_str = Py_None;
-            Py_INCREF(conversion_str);
+            conversion_str = Py_NewRef(Py_None);
         }
         else
             conversion_str = PyUnicode_FromKindAndData(PyUnicode_4BYTE_KIND,
@@ -1121,8 +1118,7 @@ formatter_parser(PyObject *ignored, PyObject *self)
         return NULL;
 
     /* take ownership, give the object to the iterator */
-    Py_INCREF(self);
-    it->str = self;
+    it->str = Py_NewRef(self);
 
     /* initialize the contained MarkupIterator */
     MarkupIterator_init(&it->it_markup, (PyObject*)self, 0, PyUnicode_GET_LENGTH(self));
@@ -1265,8 +1261,7 @@ formatter_field_name_split(PyObject *ignored, PyObject *self)
 
     /* take ownership, give the object to the iterator.  this is
        just to keep the field_name alive */
-    Py_INCREF(self);
-    it->str = self;
+    it->str = Py_NewRef(self);
 
     /* Pass in auto_number = NULL. We'll return an empty string for
        first_obj in that case. */
diff --git a/contrib/tools/python3/src/Objects/stringlib/unicodedefs.h b/contrib/tools/python3/src/Objects/stringlib/unicodedefs.h
deleted file mode 100644
index ba2ce0aeea..0000000000
--- a/contrib/tools/python3/src/Objects/stringlib/unicodedefs.h
+++ /dev/null
@@ -1,32 +0,0 @@
-#ifndef STRINGLIB_UNICODEDEFS_H
-#define STRINGLIB_UNICODEDEFS_H
-
-/* this is sort of a hack.  there's at least one place (formatting
-   floats) where some stringlib code takes a different path if it's
-   compiled as unicode. */
-#define STRINGLIB_IS_UNICODE     1
-
-#define FASTSEARCH               fastsearch
-#define STRINGLIB(F)             stringlib_##F
-#define STRINGLIB_OBJECT         PyUnicodeObject
-#define STRINGLIB_SIZEOF_CHAR    Py_UNICODE_SIZE
-#define STRINGLIB_CHAR           Py_UNICODE
-#define STRINGLIB_TYPE_NAME      "unicode"
-#define STRINGLIB_PARSE_CODE     "U"
-#define STRINGLIB_ISSPACE        Py_UNICODE_ISSPACE
-#define STRINGLIB_ISLINEBREAK    BLOOM_LINEBREAK
-#define STRINGLIB_ISDECIMAL      Py_UNICODE_ISDECIMAL
-#define STRINGLIB_TODECIMAL      Py_UNICODE_TODECIMAL
-#define STRINGLIB_STR            PyUnicode_AS_UNICODE
-#define STRINGLIB_LEN            PyUnicode_GET_SIZE
-#define STRINGLIB_NEW            PyUnicode_FromUnicode
-#define STRINGLIB_CHECK          PyUnicode_Check
-#define STRINGLIB_CHECK_EXACT    PyUnicode_CheckExact
-#define STRINGLIB_MUTABLE 0
-
-#define STRINGLIB_TOSTR          PyObject_Str
-#define STRINGLIB_TOASCII        PyObject_ASCII
-
-#define STRINGLIB_WANT_CONTAINS_OBJ 1
-
-#endif /* !STRINGLIB_UNICODEDEFS_H */