fix: zero-width word paste can be wrong in default repl

Signed-off-by: yihong0618 <[email protected]>

Author: yihong0618

Lib/_pyrepl/utils.py

@@ -63,6 +63,9 @@ class ColorSpan(NamedTuple):
 def str_width(c: str) -> int:
     if ord(c) < 128:
         return 1
+    # gh-139246 for zero-width joiner and combining characters
+    if unicodedata.combining(c) or unicodedata.category(c) == "Cf":
+        return 0
     w = unicodedata.east_asian_width(c)
     if w in ("N", "Na", "H", "A"):
         return 1

Lib/test/test_pyrepl/test_utils.py

@@ -5,10 +5,28 @@
 
 class TestUtils(TestCase):
     def test_str_width(self):
-        characters = ['a', '1', '_', '!', '\x1a', '\u263A', '\uffb9']
+        characters = [
+            'a',
+            '1',
+            '_',
+            '!',
+            '\x1a',
+            '\u263A',
+            '\uffb9',
+            '\N{LATIN SMALL LETTER E WITH ACUTE}',  # é
+            '\N{LATIN SMALL LETTER E WITH CEDILLA}', # ȩ
+        ]
         for c in characters:
             self.assertEqual(str_width(c), 1)
 
+        zero_width_characters = [
+            '\N{COMBINING ACUTE ACCENT}',
+            '\N{ZERO WIDTH JOINER}',
+        ]
+        for c in zero_width_characters:
+            with self.subTest(character=c):
+                self.assertEqual(str_width(c), 0)
+
         characters = [chr(99989), chr(99999)]
         for c in characters:
             self.assertEqual(str_width(c), 2)
@@ -25,6 +43,8 @@ def test_wlen(self):
 
         self.assertEqual(wlen('hello'), 5)
         self.assertEqual(wlen('hello' + '\x1a'), 7)
+        self.assertEqual(wlen('e\N{COMBINING ACUTE ACCENT}'), 1)
+        self.assertEqual(wlen('a\N{ZERO WIDTH JOINER}b'), 2)
 
     def test_prev_next_window(self):
         def gen_normal():

Misc/NEWS.d/next/Library/2025-09-23-09-46-46.gh-issue-139246.pzfM-w.rst

@@ -0,0 +1 @@
+fix: paste zero-width in default repl width is wrong.