Merge branch 'main' into fix-issue-141938

Author: TotalyEnglizLitrate

Lib/encodings/punycode.py

@@ -17,7 +17,7 @@ def segregate(str):
         else:
             extended.add(c)
     extended = sorted(extended)
-    return bytes(base), extended
+    return base.take_bytes(), extended
 
 def selective_len(str, max):
     """Return the length of str, considering only characters below max."""
@@ -83,7 +83,7 @@ def generate_generalized_integer(N, bias):
         t = T(j, bias)
         if N < t:
             result.append(digits[N])
-            return bytes(result)
+            return result.take_bytes()
         result.append(digits[t + ((N - t) % (36 - t))])
         N = (N - t) // (36 - t)
         j += 1
@@ -112,7 +112,7 @@ def generate_integers(baselen, deltas):
         s = generate_generalized_integer(delta, bias)
         result.extend(s)
         bias = adapt(delta, points==0, baselen+points+1)
-    return bytes(result)
+    return result.take_bytes()
 
 def punycode_encode(text):
     base, extended = segregate(text)

Lib/re/_compiler.py

@@ -375,7 +375,7 @@ def _optimize_charset(charset, iscased=None, fixup=None, fixes=None):
     # less significant byte is a bit index in the chunk (just like the
     # CHARSET matching).
 
-    charmap = bytes(charmap) # should be hashable
+    charmap = charmap.take_bytes() # should be hashable
     comps = {}
     mapping = bytearray(256)
     block = 0

Misc/NEWS.d/next/Library/2025-11-25-23-29-08.gh-issue-141968.0JnjXf.rst

@@ -0,0 +1,2 @@
+Remove data copy from :mod:`codecs` ``punycode`` encoding by using
+:meth:`bytearray.take_bytes`.

Misc/NEWS.d/next/Library/2025-11-26-14-20-10.gh-issue-141968.W139Pv.rst

@@ -0,0 +1,2 @@
+Remove data copy from :mod:`re` compilation of regexes with large charsets
+by using :meth:`bytearray.take_bytes`.

Tools/c-analyzer/cpython/_parser.py

@@ -114,7 +114,10 @@ def format_tsv_lines(lines):
     ('*', './Include/internal/mimalloc'),
 
     ('Modules/_decimal/**/*.c', 'Modules/_decimal/libmpdec'),
+
     ('Modules/_elementtree.c', 'Modules/expat'),
+    ('Modules/pyexpat.c', 'Modules/expat'),
+
     ('Modules/_hacl/*.c', 'Modules/_hacl/include'),
     ('Modules/_hacl/*.c', 'Modules/_hacl/'),
     ('Modules/_hacl/*.h', 'Modules/_hacl/include'),
@@ -125,6 +128,7 @@ def format_tsv_lines(lines):
     ('Modules/sha3module.c', 'Modules/_hacl/include'),
     ('Modules/blake2module.c', 'Modules/_hacl/include'),
     ('Modules/hmacmodule.c', 'Modules/_hacl/include'),
+
     ('Objects/stringlib/*.h', 'Objects'),
 
     # possible system-installed headers, just in case