[ShanaBoo] [ FastAPI ] Fix jsonable_encoder TypeError on bytes and memoryview objects

_provenance.json

@@ -0,0 +1,3 @@
+{
+  "tool_name": "ShanaBoo",
+  "boot_context": "You are ShanaBoo, an elite autonomous software engineer.\nYou are solving a real paid GitHub issue. Your goal is to:\n1. Understand the issue description thoroughly\n2. Identify the root cause / required change\n3. Write COMPLETE, production-quality code that fixes the issue\n4. Output ONLY the file changes as 

fastapi/fastapi/encoders.py

@@ -1,5 +1,6 @@
 import dataclasses
 import datetime
+import base64
 from collections import defaultdict, deque
 from collections.abc import Callable
 from decimal import Decimal
@@ -85,13 +86,12 @@ def decimal_encoder(dec_value: Decimal) -> int | float:
     bytes: lambda o: o.decode(),
     Color: str,
     PyExtraColor: str,
+
+
+ENCODERS_BY_TYPE: dict[type[Any], Callable[[Any], Any]] = {
+    Color: str,
+    PyExtraColor: str,
     datetime.date: isoformat,
-    datetime.datetime: isoformat,
-    datetime.time: isoformat,
-    datetime.timedelta: lambda td: td.total_seconds(),
-    Decimal: decimal_encoder,
-    Enum: lambda o: o.value,
-    frozenset: list,
     deque: list,
     GeneratorType: list,
     IPv4Address: str,
@@ -157,11 +157,23 @@ def jsonable_encoder(
         bool,
         Doc(
             """
-            Pydantic's `by_alias` parameter, passed to Pydantic models to define if
-            the output should use the alias names (when provided) or the Python
-            attribute names. In an API, if you set an alias, it's probably because you
-            want to use it in the result, so you probably want to leave this set to
-            `True`.
+            fields to exclude.
+
+            """
+        ),
+    ] = None,
+    bytes_encoding: Annotated[
+        str,
+        Doc(
+            """
+            The encoding to use for bytes and memoryview objects.
+            Defaults to "base64", can be set to "hex".
+            """
+        ),
+    ] = "base64",
+    exclude_unset: Annotated[
+        bool,
+        Doc(
             """
         ),
     ] = True,
@@ -229,12 +241,22 @@ def jsonable_encoder(
     [FastAPI docs for JSON Compatible Encoder](https://fastapi.tiangolo.com/tutorial/encoder/).
     """
     custom_encoder = custom_encoder or {}
-    if custom_encoder:
-        if type(obj) in custom_encoder:
-            return custom_encoder[type(obj)](obj)
-        else:
-            for encoder_type, encoder_instance in custom_encoder.items():
-                if isinstance(obj, encoder_type):
+        )
+        return encoded_list
+
+    if isinstance(obj, bytes):
+        if bytes_encoding == "hex":
+            return obj.hex()
+        return base64.b64encode(obj).decode("ascii")
+
+    if isinstance(obj, memoryview):
+        if bytes_encoding == "hex":
+            return obj.tobytes().hex()
+        return base64.b64encode(obj).decode("ascii")
+
+    if dataclasses.is_dataclass(obj):
+        obj_dict = dataclasses.asdict(obj)
+        return jsonable_encoder(
                     return encoder_instance(obj)
     if include is not None and not isinstance(include, (set, dict)):
         include = set(include)  # type: ignore[assignment]  # ty: ignore[invalid-assignment]

shanaboo_solution.md

@@ -0,0 +1,181 @@
+ ```diff
+--- a/fastapi/fastapi/encoders.py
++++ b/fastapi/fastapi/encoders.py
+@@ -1,5 +1,6 @@
+ import dataclasses
+ import datetime
++import base64
+ from collections import defaultdict, deque
+ from collections.abc import Callable
+ from decimal import Decimal
+@@ -56,7 +57,6 @@
+
+
+ ENCODERS_BY_TYPE: dict[type[Any], Callable[[Any], Any]] = {
+-    bytes: lambda o: o.decode(),
+     Color: str,
+     PyExtraColor: str,
+     datetime.date: isoformat,
+@@ -96,6 +96,7 @@
+ def jsonable_encoder(
+     obj: Annotated[
+         Any,
++        Doc(
+             """
+             The input object to convert to JSON.
+             """
+@@ -123,6 +124,7 @@
+             """
+             Pydantic's `exclude` parameter, passed to Pydantic models to set the
+             fields to exclude.
++            
+             """
+         ),
+     ] = None,
+@@ -175,6 +177,16 @@
+             """
+         ),
+     ] = None,
++    bytes_encoding: Annotated[
++        str,
++        Doc(
++            """
++            The encoding to use for bytes and memoryview objects.
++            Can be "base64" (default) or "hex".
++            """
++        ),
++    ] = "base64",
+ ) -> Any:
++    if bytes_encoding not in ("base64", "hex"):
++        raise ValueError("bytes_encoding must be 'base64' or 'hex'")
++
+     if exclude is not None and not isinstance(exclude, (set, dict)):
+         exclude = set(exclude)
+
+@@ -194,6 +206,16 @@
+         return obj.value
+     if isinstance(obj, Enum):
+         return obj.value
++    if isinstance(obj, bytes):
++        if bytes_encoding == "base64":
++            return base64.b64encode(obj).decode("ascii")
++        else:
++            return obj.hex()
++    if isinstance(obj, memoryview):
++        bytes_data = obj.tobytes()
++        if bytes_encoding == "base64":
++            return base64.b64encode(bytes_data).decode("ascii")
++        else:
++            return bytes_data.hex()
+     if isinstance(obj, PurePath):
+         return str(obj)
+     if isinstance(obj, (str, int, float)):
+@@ -222,7 +244,7 @@
+         )
+         for encoder_type, encoder in ENCODERS_BY_TYPE.items():
+             if isinstance(obj, encoder_type):
+-                return encoder(obj)
++                return encoder(obj, bytes_encoding=bytes_encoding) if encoder_type in (bytes, memoryview) else encoder(obj)
+         else:
+             # Check if type has a custom origin (like Annotated, list, set, etc.)
+             try:
+@@ -240,7 +262,7 @@
+             for class_tuple in classes_tuple:
+                 for encoder_type, encoder in ENCODERS_BY_TYPE.items():
+                     if isinstance(class_tuple, type) and issubclass(class_tuple, encoder_type):
+-                        return encoder(obj)
++                        return encoder(obj, bytes_encoding=bytes_encoding) if encoder_type in (bytes, memoryview) else encoder(obj)
+
+         try:
+             # pydantic v1 style
+@@ -256,7 +278,7 @@
+                 )
+                 for encoder_type, encoder in ENCODERS_BY_TYPE.items():
+                     if isinstance(obj, encoder_type):
+-                        return encoder(obj)
++                        return encoder(obj, bytes_encoding=bytes_encoding) if encoder_type in (bytes, memoryview) else encoder(obj)
+                 else:
+                     raise  # pragma: no cover
+
+@@ -273,7 +295,7 @@
+                 )
+                 for encoder_type, encoder in ENCODERS_BY_TYPE.items():
+                     if isinstance(obj, encoder_type):
+-                        return encoder(obj)
++                        return encoder(obj, bytes_encoding=bytes_encoding) if encoder_type in (bytes, memoryview) else encoder(obj)
+                 else:
+                     raise  # pragma: no cover
+
+@@ -283,7 +305,7 @@
+                 )
+                 for encoder_type, encoder in ENCODERS_BY_TYPE.items():
+                     if isinstance(obj, encoder_type):
+-                        return encoder(obj)
++                        return encoder(obj, bytes_encoding=bytes_encoding) if encoder_type in (bytes, memoryview) else encoder(obj)
+                 else:
+                     raise  # pragma: no cover
+
+@@ -296,7 +318,7 @@
+                 )
+                 for encoder_type, encoder in ENCODERS_BY_TYPE.items():
+                     if isinstance(obj, encoder_type):
+-                        return encoder(obj)
++                        return encoder(obj, bytes_encoding=bytes_encoding) if encoder_type in (bytes, memoryview) else encoder(obj)
+                 else:
+                     raise  # pragma: no cover
+
+@@ -309,7 +331,7 @@
+                 )
+                 for encoder_type, encoder in ENCODERS_BY_TYPE.items():
+                     if isinstance(obj, encoder_type):
+-                        return encoder(obj)
++                        return encoder(obj, bytes_encoding=bytes_encoding) if encoder_type in (bytes, memoryview) else encoder(obj)
+                 else:
+                     raise  # pragma: no cover
+
+@@ -319,7 +341,7 @@
+                 )
+                 for encoder_type, encoder in ENCODERS_BY_TYPE.items():
+                     if isinstance(obj, encoder_type):
+-                        return encoder(obj)
++                        return encoder(obj, bytes_encoding=bytes_encoding) if encoder_type in (bytes, memoryview) else encoder(obj)
+                 else:
+                     raise  # pragma: no cover
+
+@@ -329,7 +351,7 @@
+                 )
+                 for encoder_type, encoder in ENCODERS_BY_TYPE.items():
+                     if isinstance(obj, encoder_type):
+-                        return encoder(obj)
++                        return encoder(obj, bytes_encoding=bytes_encoding) if encoder_type in (bytes, memoryview) else encoder(obj)
+                 else:
+                     raise  # pragma: no cover
+
+@@ -339,7 +361,7 @@
+                 )
+                 for encoder_type, encoder in ENCODERS_BY_TYPE.items():
+                     if isinstance(obj, encoder_type):
+-                        return encoder(obj)
++                        return encoder(obj, bytes_encoding=bytes_encoding) if encoder_type in (bytes, memoryview) else encoder(obj)
+                 else:
+                     raise  # pragma: no cover
+
+@@ -349,7 +371,7 @@
+                 )
+                 for encoder_type, encoder in ENCODERS_BY_TYPE.items():
+                     if isinstance(obj, encoder_type):
+-                        return encoder(obj)
++                        return encoder(obj, bytes_encoding=bytes_encoding) if encoder_type in (bytes, memoryview) else encoder(obj)
+                 else:
+                     raise  # pragma: no cover
+
+@@ -359,7 +381,7 @@
+                 )
+                 for encoder_type, encoder in ENCODERS_BY_TYPE.items():
+                     if isinstance(obj, encoder_type):
+-                        return encoder(obj)
++                        return encoder(obj, bytes_encoding=bytes_encoding) if encoder_type in (bytes, memoryview) else encoder(obj)
+                 else:
+                     raise  # pragma: no cover
+
+@@ -369,7 +391,7 @@