ast.unparse is not working correctly for strings with multiple quotes

[sunmy2019]

Bug report

CPython versions tested on:

3.11

Operating systems tested on:

macOS

A clear and concise description of the bug:

During the implementation of #108553, I discovered this bug in the original implementation.

Due to the new syntax, the fix for 3.12 above cannot be reused for 3.11 and below.

import ast

s = "f\"'''{1}\\\"\\\"\\\"\" "
print(s)
eval(s)

s = ast.parse(s)
s1 = ast.unparse(s)
print(s1)
eval(s1)

output:

f"'''{1}\"\"\"" 
f''''{1}"""'
Traceback (most recent call last):
  File "/Users/xxx/work/cpython/a.py", line 10, in <module>
    eval(s1)
  File "<string>", line 1
    f''''{1}"""'
    ^
SyntaxError: unterminated triple-quoted string literal (detected at line 1)

Linked PRs

• [3.11] gh-108843: fix ast.unparse for f-string with many quotes #108980
• gh-108843: fix ast.unparse for f-string with many quotes #108981
• [3.12] gh-108843: fix ast.unparse for f-string with many quotes (GH-108981) #109541

[hauntsaninja]

Nice spot, I've opened a PR that fixes on 3.11 over here: #108980

[hauntsaninja]

@sunmy2019 oh hmm, I got a little confused by the history here, looks like this isn't actually fixed on main?

A similar patch should make it work on main, something like the following:

diff --git a/Lib/ast.py b/Lib/ast.py
index 17ec7ff6f8..f706d1f282 100644
--- a/Lib/ast.py
+++ b/Lib/ast.py
@@ -1236,17 +1236,35 @@ def visit_JoinedStr(self, node):
 
         new_fstring_parts = []
         quote_types = list(_ALL_QUOTES)
+        fallback_to_repr = False
         for value, is_constant in fstring_parts:
             if is_constant:
-                value, quote_types = self._str_literal_helper(
+                value, new_quote_types = self._str_literal_helper(
                     value,
                     quote_types=quote_types,
                     escape_special_whitespace=True,
                 )
+                if set(new_quote_types).isdisjoint(quote_types):
+                    fallback_to_repr = True
+                    break
+                quote_types = new_quote_types
             elif "\n" in value:
                 quote_types = [q for q in quote_types if q in _MULTI_QUOTES]
             new_fstring_parts.append(value)
 
+        if fallback_to_repr:
+            # If we weren't able to find a quote type that works for all parts
+            # of the JoinedStr, fallback to using repr and triple single quotes.
+            quote_types = ["'''"]
+            new_fstring_parts.clear()
+            for value, is_constant in fstring_parts:
+                if is_constant:
+                    value = repr("'" + '"' + value)  # force repr to escape all quotes
+                    expected_prefix = r"'\'" + '"'
+                    assert value.startswith(expected_prefix), repr(value)
+                    value = value[len(expected_prefix):-1]
+                new_fstring_parts.append(value)
+
         value = "".join(new_fstring_parts)
         quote_type = quote_types[0]
         self.write(f"{quote_type}{value}{quote_type}")
diff --git a/Lib/test/test_unparse.py b/Lib/test/test_unparse.py
index 38c59e6d43..c09865894d 100644
--- a/Lib/test/test_unparse.py
+++ b/Lib/test/test_unparse.py
@@ -635,6 +635,16 @@ def test_star_expr_assign_target_multiple(self):
         self.check_src_roundtrip("[a, b] = [c, d] = [e, f] = g")
         self.check_src_roundtrip("a, b = [c, d] = e, f = g")
 
+    def test_multiquote_joined_string(self):
+        self.check_ast_roundtrip("f\"'''{1}\\\"\\\"\\\"\" ")
+        self.check_ast_roundtrip("""f"'''{1}""\\"" """)
+        self.check_ast_roundtrip("""f'""\"{1}''' """)
+        self.check_ast_roundtrip("""f'""\"{1}""\\"' """)
+
+        self.check_ast_roundtrip("""f"'''{"\\n"}""\\"" """)
+        self.check_ast_roundtrip("""f'""\"{"\\n"}''' """)
+        self.check_ast_roundtrip("""f'""\"{"\\n"}""\\"' """)
+
 
 class ManualASTCreationTestCase(unittest.TestCase):
     """Test that AST nodes created without a type_params field unparse correctly."""

Should I open a PR?

[sunmy2019]

I got a little confused by the history here, looks like this isn't actually fixed on main?

Yes. I got it reverted as requested to separate the fix out.

Should I open a PR?

Go ahead!