Sync pre-release CPython 3.15 branch from GitHub (2026-08-26)

Summary:
Imported python/cpython `3.15.0rc1+dev` from upstream rev [`1c826a4`](https://www.github.com/python/cpython/commit/1c826a40a2f9606e284fd3d6d5f7c7c6732327ed) (committed 2026-08-26 00:59:49+00:00).

# Commit Info

- Base: (`3.15.0rc1+dev`) - [`c32b891`](https://www.github.com/python/cpython/commit/c32b891de6421e7943eec441144f14986af07507) (commit date: 2026-08-25 00:39:52+00:00)
- Imported: (`3.15.0rc1+dev`) - [`1c826a4`](https://www.github.com/python/cpython/commit/1c826a40a2f9606e284fd3d6d5f7c7c6732327ed) (commit date: 2026-08-26 00:59:49+00:00)

# Noteworthy file changes

- Low-signal files (2 added) (NEWS.d, docs, .github)

Complete list of added/removed files: https://www.internalfb.com/intern/everpaste/?color=0&handle=GD2foS7Vl3rB6xoIAPXuY5YEhq4Dbr0LAAAz

Reviewed By: itamaro

Differential Revision: D117473160

fbshipit-source-id: 84aaf4643c242744e7a5b9772cc796fc89da7327

Author: generatedunixname1734921407115435

.github/workflows/build.yml

@@ -332,13 +332,13 @@ jobs:
           # unsupported as it most resembles other 1.1.1-work-a-like ssl APIs
           # supported by important vendors such as AWS-LC.
           - { name: openssl, version: 1.1.1w }
-          - { name: openssl, version: 3.0.21 }
-          - { name: openssl, version: 3.4.6 }
-          - { name: openssl, version: 3.5.7 }
-          - { name: openssl, version: 3.6.3 }
-          - { name: openssl, version: 4.0.1 }
+          - { name: openssl, version: 3.0.22 }
+          - { name: openssl, version: 3.4.7 }
+          - { name: openssl, version: 3.5.8 }
+          - { name: openssl, version: 3.6.4 }
+          - { name: openssl, version: 4.0.2 }
           ## AWS-LC
-          - { name: aws-lc, version: 5.0.0 }
+          - { name: aws-lc, version: 5.5.0 }
     env:
       SSLLIB_VER: ${{ matrix.ssllib.version }}
       MULTISSL_DIR: ${{ github.workspace }}/multissl
@@ -443,7 +443,7 @@ jobs:
     needs: build-context
     if: needs.build-context.outputs.run-ubuntu == 'true'
     env:
-      OPENSSL_VER: 3.5.7
+      OPENSSL_VER: 3.5.8
       PYTHONSTRICTEXTENSIONBUILD: 1
     steps:
     - uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
@@ -551,7 +551,7 @@ jobs:
       matrix:
         os: [ubuntu-26.04]
     env:
-      OPENSSL_VER: 3.5.7
+      OPENSSL_VER: 3.5.8
       PYTHONSTRICTEXTENSIONBUILD: 1
       ASAN_OPTIONS: detect_leaks=0:allocator_may_return_null=1:handle_segv=0
     steps:

.github/workflows/reusable-san.yml

@@ -17,7 +17,7 @@ permissions:
 
 env:
   FORCE_COLOR: 1
-  OPENSSL_VER: 3.5.7
+  OPENSSL_VER: 3.5.8
 
 jobs:
   build-san-reusable:

.github/workflows/reusable-ubuntu.yml

@@ -35,7 +35,7 @@ jobs:
     runs-on: ${{ inputs.os }}
     timeout-minutes: 60
     env:
-      OPENSSL_VER: 3.5.7
+      OPENSSL_VER: 3.5.8
       PYTHONSTRICTEXTENSIONBUILD: 1
       TERM: linux
     steps:

Doc/library/csv.rst

@@ -369,6 +369,8 @@ The :mod:`!csv` module defines the following constants:
    Instructs :class:`writer` objects to only quote those fields which contain
    special characters such as *delimiter*, *quotechar*, ``'\r'``, ``'\n'``
    or any of the characters in *lineterminator*.
+   If *doublequote* is :const:`False` and *escapechar* is set,
+   the *quotechar* is escaped instead of causing the field to be quoted.
 
 
 .. data:: QUOTE_NONNUMERIC

Doc/library/stdtypes.rst

@@ -2594,7 +2594,8 @@ expression support in the :mod:`re` module).
 
    Return a list of the words in the string, using *sep* as the delimiter string.
    If *maxsplit* is given, at most *maxsplit* splits are done, the *rightmost*
-   ones.  If *sep* is not specified or ``None``, any whitespace string is a
+   ones.  If *sep* is not specified or ``None``, any
+   :meth:`whitespace <str.isspace>` string is a
    separator.  Except for splitting from the right, :meth:`rsplit` behaves like
    :meth:`split` which is described in detail below.
 
@@ -2654,7 +2655,8 @@ expression support in the :mod:`re` module).
       ['1', '2', '3<4']
 
    If *sep* is not specified or is ``None``, a different splitting algorithm is
-   applied: runs of consecutive whitespace are regarded as a single separator,
+   applied: runs of consecutive :meth:`whitespace <str.isspace>` are regarded
+   as a single separator,
    and the result will contain no empty strings at the start or end if the
    string has leading or trailing whitespace.  Consequently, splitting an empty
    string or a string consisting of just whitespace with a ``None`` separator
@@ -3920,7 +3922,8 @@ produce new objects.
    Return a copy of the sequence with specified leading bytes removed.  The
    *bytes* argument is a binary sequence specifying the set of byte values to
    be removed.  If omitted or ``None``, the *bytes* argument defaults
-   to removing ASCII whitespace.  The *bytes* argument is not a prefix;
+   to removing :meth:`ASCII whitespace <bytes.isspace>`.
+   The *bytes* argument is not a prefix;
    rather, all combinations of its values are stripped::
 
       >>> b'   spacious   '.lstrip()
@@ -3964,7 +3967,8 @@ produce new objects.
    Split the binary sequence into subsequences of the same type, using *sep*
    as the delimiter string. If *maxsplit* is given, at most *maxsplit* splits
    are done, the *rightmost* ones.  If *sep* is not specified or ``None``,
-   any subsequence consisting solely of ASCII whitespace is a separator.
+   any subsequence consisting solely of
+   :meth:`ASCII whitespace <bytes.isspace>` is a separator.
    Except for splitting from the right, :meth:`rsplit` behaves like
    :meth:`split` which is described in detail below.
 
@@ -3975,7 +3979,8 @@ produce new objects.
    Return a copy of the sequence with specified trailing bytes removed.  The
    *bytes* argument is a binary sequence specifying the set of byte values to
    be removed.  If omitted or ``None``, the *bytes* argument defaults to
-   removing ASCII whitespace.  The *bytes* argument is not a suffix; rather,
+   removing :meth:`ASCII whitespace <bytes.isspace>`.
+   The *bytes* argument is not a suffix; rather,
    all combinations of its values are stripped::
 
       >>> b'   spacious   '.rstrip()
@@ -4028,7 +4033,8 @@ produce new objects.
       [b'1', b'2', b'3<4']
 
    If *sep* is not specified or is ``None``, a different splitting algorithm
-   is applied: runs of consecutive ASCII whitespace are regarded as a single
+   is applied: runs of consecutive :meth:`ASCII whitespace <bytes.isspace>`
+   are regarded as a single
    separator, and the result will contain no empty strings at the start or
    end if the sequence has leading or trailing whitespace.  Consequently,
    splitting an empty sequence or a sequence consisting solely of ASCII
@@ -4051,7 +4057,8 @@ produce new objects.
    Return a copy of the sequence with specified leading and trailing bytes
    removed. The *bytes* argument is a binary sequence specifying the set of
    byte values to be removed.  If omitted or ``None``, the *bytes*
-   argument defaults to removing ASCII whitespace. The *bytes* argument is
+   argument defaults to removing :meth:`ASCII whitespace <bytes.isspace>`.
+   The *bytes* argument is
    not a prefix or suffix; rather, all combinations of its values are
    stripped::
 

Doc/whatsnew/3.15.rst

@@ -1001,7 +1001,7 @@ base64
 
 * Added the *canonical* parameter in
   :func:`~base64.b32decode`, :func:`~base64.b32hexdecode`,
-  :func:`~base64.b64decode`, :func:`~base64.urlsafe_b64decode`,
+  :func:`~base64.b64decode`,
   :func:`~base64.a85decode`, :func:`~base64.b85decode`, and
   :func:`~base64.z85decode`,
   to reject encodings with non-zero padding bits or other non-canonical

Lib/profiling/sampling/_flamegraph_assets/flamegraph.js

@@ -955,8 +955,8 @@ function populateProfileSummary(data) {
     : (stats.total_samples ?? data.value ?? 0);
   const duration = stats.duration_sec || 0;
   const sampleRate = stats.sample_rate || (duration > 0 ? totalSamples / duration : 0);
-  const errorRate = stats.error_rate || 0;
-  const missedSamples= stats.missed_samples || 0;
+  const errorRate = stats.error_rate;
+  const missedSamples = stats.missed_samples;
 
   const samplesEl = document.getElementById('stat-total-samples');
   if (samplesEl) samplesEl.textContent = formatNumber(totalSamples);

Lib/profiling/sampling/stack_collector.py

@@ -212,7 +212,7 @@ def _get_module_name(self, filename, path_info):
             self._module_cache[filename] = module_name
         return module_name
 
-    def _convert_to_flamegraph_format(self):
+    def _convert_to_flamegraph_format(self, *, min_samples=None):
         if self._total_samples == 0:
             return {
                 "name": self._string_table.intern("No Data"),
@@ -302,7 +302,8 @@ def convert_children(children, min_samples, path_info):
 
         # Filter out very small functions (less than 0.1% of total samples)
         total_samples = self._total_samples
-        min_samples = max(1, int(total_samples * 0.001))
+        if min_samples is None:
+            min_samples = max(1, int(total_samples * 0.001))
         path_info = get_python_path_info()
 
         root_children = convert_children(self._root["children"], min_samples, path_info)
@@ -690,7 +691,15 @@ def _add_elided_flamegraph(self, current_flamegraph, current_stats, baseline_sta
         """Calculate elided paths and add elided flamegraph to stats."""
         self._elided_paths = baseline_stats.keys() - current_stats.keys()
 
-        current_flamegraph["stats"]["elided_count"] = len(self._elided_paths)
+        # A sampled stack can end at an internal path that also has elided
+        # descendants.  Count every disappeared path with self samples, not
+        # just the leaves of the elided path tree.
+        elided_stacks = {
+            path
+            for path in self._elided_paths
+            if baseline_stats[path]["self"] > 0
+        }
+        current_flamegraph["stats"]["elided_count"] = len(elided_stacks)
 
         if self._elided_paths:
             elided_flamegraph = self._build_elided_flamegraph(baseline_stats, scale)
@@ -713,7 +722,9 @@ def _build_elided_flamegraph(self, baseline_stats, scale):
         orig_get_source = self._baseline_collector._get_source_lines
         self._baseline_collector._get_source_lines = lambda func: None
         try:
-            baseline_data = self._baseline_collector._convert_to_flamegraph_format()
+            baseline_data = self._baseline_collector._convert_to_flamegraph_format(
+                min_samples=1
+            )
         finally:
             self._baseline_collector._get_source_lines = orig_get_source
 

Lib/test/test_profiling/test_sampling_profiler/test_collectors.py

@@ -1665,7 +1665,7 @@ def test_diff_flamegraph_elided_stacks(self):
 
         data = diff._convert_to_flamegraph_format()
 
-        self.assertGreater(data["stats"]["elided_count"], 0)
+        self.assertEqual(data["stats"]["elided_count"], 1)
         self.assertIn("elided_flamegraph", data["stats"])
         elided = data["stats"]["elided_flamegraph"]
         self.assertTrue(elided["stats"]["is_differential"])
@@ -1681,6 +1681,74 @@ def test_diff_flamegraph_elided_stacks(self):
         self.assertGreater(child["baseline"], 0)
         self.assertAlmostEqual(child["diff"], -child["baseline"])
 
+    def test_diff_flamegraph_counts_elided_stacks_not_paths(self):
+        """Internal and leaf stack endings are counted separately."""
+        internal_stack = [
+            MockInterpreterInfo(0, [
+                MockThreadInfo(1, [
+                    MockFrameInfo("file.py", 20, "old_mid"),
+                    MockFrameInfo("file.py", 10, "root"),
+                ])
+            ])
+        ]
+        leaf_stack = [
+            MockInterpreterInfo(0, [
+                MockThreadInfo(1, [
+                    MockFrameInfo("file.py", 30, "old_leaf"),
+                    MockFrameInfo("file.py", 20, "old_mid"),
+                    MockFrameInfo("file.py", 10, "root"),
+                ])
+            ])
+        ]
+        current_frames = [
+            MockInterpreterInfo(0, [
+                MockThreadInfo(1, [MockFrameInfo("file.py", 10, "root")])
+            ])
+        ]
+
+        diff = make_diff_collector_with_mock_baseline(
+            [internal_stack, leaf_stack]
+        )
+        diff.collect(current_frames)
+
+        data = diff._convert_to_flamegraph_format()
+        self.assertEqual(data["stats"]["elided_count"], 2)
+
+    def test_diff_flamegraph_renders_small_elided_stack(self):
+        """Elided stacks are not removed by the significance filter."""
+        common_frames = [
+            MockInterpreterInfo(0, [
+                MockThreadInfo(1, [
+                    MockFrameInfo("file.py", 20, "common"),
+                    MockFrameInfo("file.py", 10, "root"),
+                ])
+            ])
+        ]
+        old_frames = [
+            MockInterpreterInfo(0, [
+                MockThreadInfo(1, [
+                    MockFrameInfo("file.py", 30, "old_tiny"),
+                    MockFrameInfo("file.py", 10, "root"),
+                ])
+            ])
+        ]
+
+        diff = make_diff_collector_with_mock_baseline(
+            [common_frames] * 1999 + [old_frames]
+        )
+        for _ in range(1999):
+            diff.collect(common_frames)
+
+        data = diff._convert_to_flamegraph_format()
+        self.assertEqual(data["stats"]["elided_count"], 1)
+        self.assertIn("elided_flamegraph", data["stats"])
+
+        elided = data["stats"]["elided_flamegraph"]
+        strings = elided["strings"]
+        self.assertIsNotNone(
+            find_child_by_name(elided.get("children", []), strings, "old_tiny")
+        )
+
     def test_diff_flamegraph_elided_top_level_root(self):
         """Elided top-level roots do not crash metadata generation."""
         baseline_frames_1 = [

Misc/NEWS.d/3.15.0b1.rst

@@ -1081,10 +1081,10 @@ Deprecate :meth:`http.cookies.Morsel.js_output` and
 .. nonce: iHWO0v
 .. section: Library
 
-Add a *canonical* keyword-only parameter to the base16, base32, base64,
-base85, ascii85, and Z85 decoders in :mod:`base64` and :mod:`binascii`. When
-true, encodings with non-zero padding bits (base16/32/64) or non-canonical
-encodings (base85/ascii85) are rejected.  Single-character final groups in
+Add a *canonical* keyword-only parameter to the base32, base64, base85,
+ascii85, and Z85 decoders in :mod:`base64` and :mod:`binascii`. When true,
+encodings with non-zero padding bits (base32/64) or non-canonical encodings
+(base85/ascii85) are rejected.  Single-character final groups in
 :func:`binascii.a2b_ascii85` and :func:`binascii.a2b_base85` are now always
 rejected as encoding violations, regardless of *canonical*; previously they
 were silently ignored and produced no output bytes.