From 10cb96b511ca2430ef0d9dcbf877d6fb306ae70f Mon Sep 17 00:00:00 2001
From: Jeff Allen <ja.py@farowl.co.uk>
Date: Sat, 23 Dec 2023 10:03:56 +0000
Subject: [PATCH 1/4] Fix GH-113028 divergence in pickling str

This fixes a divergence between the Python and C implementations of pickle
for protocol 0, such that it pickle.py fails to re-use the first pickled
representation of strings involving characters that have to be escaped.
---
 Lib/pickle.py            | 14 +++++++-------
 Lib/test/pickletester.py |  8 ++++++++
 2 files changed, 15 insertions(+), 7 deletions(-)

diff --git a/Lib/pickle.py b/Lib/pickle.py
index 4f5ad5b71e8899..988c0887341310 100644
--- a/Lib/pickle.py
+++ b/Lib/pickle.py
@@ -857,13 +857,13 @@ def save_str(self, obj):
             else:
                 self.write(BINUNICODE + pack("<I", n) + encoded)
         else:
-            obj = obj.replace("\\", "\\u005c")
-            obj = obj.replace("\0", "\\u0000")
-            obj = obj.replace("\n", "\\u000a")
-            obj = obj.replace("\r", "\\u000d")
-            obj = obj.replace("\x1a", "\\u001a")  # EOF on DOS
-            self.write(UNICODE + obj.encode('raw-unicode-escape') +
-                       b'\n')
+            # Escape what raw-unicode-escape doesn't, but memoize the original.
+            tmp = obj.replace("\\", "\\u005c")
+            tmp = tmp.replace("\0", "\\u0000")
+            tmp = tmp.replace("\n", "\\u000a")
+            tmp = tmp.replace("\r", "\\u000d")
+            tmp = tmp.replace("\x1a", "\\u001a")  # EOF on DOS
+            self.write(UNICODE + tmp.encode('raw-unicode-escape') + b'\n')
         self.memoize(obj)
     dispatch[str] = save_str
 
diff --git a/Lib/test/pickletester.py b/Lib/test/pickletester.py
index fd446c8145850c..bd5660b63fab54 100644
--- a/Lib/test/pickletester.py
+++ b/Lib/test/pickletester.py
@@ -1825,6 +1825,14 @@ def test_unicode_high_plane(self):
             t2 = self.loads(p)
             self.assert_is_copy(t, t2)
 
+    def test_unicode_memoization(self):
+        # gh-
+        for proto in protocols:
+            for s in '', 'xyz', 'xyz\n', 'x\\yz', 'x\xa1yz\r':
+                p = self.dumps((s, s), proto)
+                s1, s2 = self.loads(p)
+                self.assertIs(s1, s2)
+
     def test_bytes(self):
         for proto in protocols:
             for s in b'', b'xyz', b'xyz'*100:

From 422f56762a446eaffce58e8021841920fb2ed898 Mon Sep 17 00:00:00 2001
From: Jeff Allen <ja.py@farowl.co.uk>
Date: Sat, 23 Dec 2023 15:35:04 +0000
Subject: [PATCH 2/4] Add comment as to intent of test

---
 Lib/test/pickletester.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/Lib/test/pickletester.py b/Lib/test/pickletester.py
index bd5660b63fab54..74b82caf742f20 100644
--- a/Lib/test/pickletester.py
+++ b/Lib/test/pickletester.py
@@ -1826,7 +1826,7 @@ def test_unicode_high_plane(self):
             self.assert_is_copy(t, t2)
 
     def test_unicode_memoization(self):
-        # gh-
+        # Repeated str is re-used (even when escapes added).
         for proto in protocols:
             for s in '', 'xyz', 'xyz\n', 'x\\yz', 'x\xa1yz\r':
                 p = self.dumps((s, s), proto)

From bf56b178ceb5e615a2bbe07cb595c9fdf1e908eb Mon Sep 17 00:00:00 2001
From: "blurb-it[bot]" <43283697+blurb-it[bot]@users.noreply.github.com>
Date: Sat, 23 Dec 2023 16:51:17 +0000
Subject: [PATCH 3/4] =?UTF-8?q?=F0=9F=93=9C=F0=9F=A4=96=20Added=20by=20blu?=
 =?UTF-8?q?rb=5Fit.?=
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

---
 .../next/Library/2023-12-23-16-51-17.gh-issue-113028.3Jmdoj.rst  | 1 +
 1 file changed, 1 insertion(+)
 create mode 100644 Misc/NEWS.d/next/Library/2023-12-23-16-51-17.gh-issue-113028.3Jmdoj.rst

diff --git a/Misc/NEWS.d/next/Library/2023-12-23-16-51-17.gh-issue-113028.3Jmdoj.rst b/Misc/NEWS.d/next/Library/2023-12-23-16-51-17.gh-issue-113028.3Jmdoj.rst
new file mode 100644
index 00000000000000..d4255b4b6cffcc
--- /dev/null
+++ b/Misc/NEWS.d/next/Library/2023-12-23-16-51-17.gh-issue-113028.3Jmdoj.rst
@@ -0,0 +1 @@
+When a second reference to a string appears in the input to `pickle`, and the Python implementation is in use, we are guaranteed that a single copy gets pickled and is shared when reloaded. Previously, in protocol 0, when a string contained certain characters (e.g. newline) it resulted in duplicate objects.

From 7e25cc85119000ba47504c7c790acffc593c2b89 Mon Sep 17 00:00:00 2001
From: Jeff Allen <ja.py@farowl.co.uk>
Date: Sat, 23 Dec 2023 17:36:18 +0000
Subject: [PATCH 4/4] Hand-edit blurb for readability and to satisfy lint.

---
 .../Library/2023-12-23-16-51-17.gh-issue-113028.3Jmdoj.rst | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/Misc/NEWS.d/next/Library/2023-12-23-16-51-17.gh-issue-113028.3Jmdoj.rst b/Misc/NEWS.d/next/Library/2023-12-23-16-51-17.gh-issue-113028.3Jmdoj.rst
index d4255b4b6cffcc..5f66d6a00b4d3d 100644
--- a/Misc/NEWS.d/next/Library/2023-12-23-16-51-17.gh-issue-113028.3Jmdoj.rst
+++ b/Misc/NEWS.d/next/Library/2023-12-23-16-51-17.gh-issue-113028.3Jmdoj.rst
@@ -1 +1,6 @@
-When a second reference to a string appears in the input to `pickle`, and the Python implementation is in use, we are guaranteed that a single copy gets pickled and is shared when reloaded. Previously, in protocol 0, when a string contained certain characters (e.g. newline) it resulted in duplicate objects.
+When a second reference to a string appears in the input to :mod:`pickle`,
+and the Python implementation is in use,
+we are guaranteed that a single copy gets pickled
+and a single object is shared when reloaded.
+Previously, in protocol 0, when a string contained certain characters
+(e.g. newline) it resulted in duplicate objects.