bpo-43323: Fix UnicodeEncodeError in the email module (GH-32137)

It was raised if the charset itself contains characters not encodable
in UTF-8 (in particular \udcxx characters representing non-decodable
bytes in the source).

2 files changed, 6 insertions, 6 deletions

diff --git a/Lib/email/_encoded_words.py b/Lib/email/_encoded_words.py
index 295ae7e..6795a60 100644
--- a/Lib/email/_encoded_words.py
+++ b/Lib/email/_encoded_words.py
@@ -179,15 +179,15 @@ def decode(ew):
     # Turn the CTE decoded bytes into unicode.
     try:
         string = bstring.decode(charset)
-    except UnicodeError:
+    except UnicodeDecodeError:
         defects.append(errors.UndecodableBytesDefect("Encoded word "
-            "contains bytes not decodable using {} charset".format(charset)))
+            f"contains bytes not decodable using {charset!r} charset"))
         string = bstring.decode(charset, 'surrogateescape')
-    except LookupError:
+    except (LookupError, UnicodeEncodeError):
         string = bstring.decode('ascii', 'surrogateescape')
         if charset.lower() != 'unknown-8bit':
-            defects.append(errors.CharsetError("Unknown charset {} "
-                "in encoded word; decoded as unknown bytes".format(charset)))
+            defects.append(errors.CharsetError(f"Unknown charset {charset!r} "
+                f"in encoded word; decoded as unknown bytes"))
     return string, charset, lang, defects
 
 
diff --git a/Lib/email/_header_value_parser.py b/Lib/email/_header_value_parser.py
index 51d355f..8a8fb8b 100644
--- a/Lib/email/_header_value_parser.py
+++ b/Lib/email/_header_value_parser.py
@@ -781,7 +781,7 @@ class MimeParameters(TokenList):
                     else:
                         try:
                             value = value.decode(charset, 'surrogateescape')
-                        except LookupError:
+                        except (LookupError, UnicodeEncodeError):
                             # XXX: there should really be a custom defect for
                             # unknown character set to make it easy to find,
                             # because otherwise unknown charset is a silent