Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

feat(core): Streaming multipart parser #3872

Merged
merged 11 commits into from
Nov 28, 2024
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions litestar/_kwargs/extractors.py
Original file line number Diff line number Diff line change
Expand Up @@ -340,8 +340,8 @@ async def _extract_multipart(
connection.scope["_form"] = form_values = ( # type: ignore[typeddict-unknown-key]
connection.scope["_form"] # type: ignore[typeddict-item]
if "_form" in connection.scope
else parse_multipart_form(
body=await connection.body(),
else await parse_multipart_form(
stream=connection.stream(),
boundary=connection.content_type[-1].get("boundary", "").encode(),
multipart_form_part_limit=multipart_form_part_limit,
type_decoders=connection.route_handler.resolve_type_decoders(),
Expand Down
169 changes: 64 additions & 105 deletions litestar/_multipart.py
Original file line number Diff line number Diff line change
@@ -1,41 +1,22 @@
"""The contents of this file were adapted from sanic.

MIT License

Copyright (c) 2016-present Sanic Community

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.
"""

from __future__ import annotations

import re
from collections import defaultdict
from email.utils import decode_rfc2231
from typing import TYPE_CHECKING, Any
from urllib.parse import unquote
from typing import TYPE_CHECKING, Any, AsyncGenerator

from multipart import ( # type: ignore[import-untyped]
MultipartSegment,
ParserError,
ParserLimitReached,
PushMultipartParser,
)

from litestar.datastructures.upload_file import UploadFile
from litestar.exceptions import ValidationException
from litestar.exceptions import ClientException

__all__ = ("parse_body", "parse_content_header", "parse_multipart_form")
__all__ = ("parse_content_header", "parse_multipart_form")

from litestar.utils.compat import async_next

if TYPE_CHECKING:
from litestar.types import TypeDecodersSequence
Expand Down Expand Up @@ -67,42 +48,16 @@
return value.strip().lower(), options


def parse_body(body: bytes, boundary: bytes, multipart_form_part_limit: int) -> list[bytes]:
"""Split the body using the boundary
and validate the number of form parts is within the allowed limit.

Args:
body: The form body.
boundary: The boundary used to separate form components.
multipart_form_part_limit: The limit of allowed form components

Returns:
A list of form components.
"""
if not (body and boundary):
return []

form_parts = body.split(boundary, multipart_form_part_limit + 3)[1:-1]

if len(form_parts) > multipart_form_part_limit:
raise ValidationException(
f"number of multipart components exceeds the allowed limit of {multipart_form_part_limit}, "
f"this potentially indicates a DoS attack"
)

return form_parts


def parse_multipart_form(
body: bytes,
async def parse_multipart_form( # noqa: C901
stream: AsyncGenerator[bytes, None],
boundary: bytes,
multipart_form_part_limit: int = 1000,
type_decoders: TypeDecodersSequence | None = None,
) -> dict[str, Any]:
"""Parse multipart form data.

Args:
body: Body of the request.
stream: Body of the request.
boundary: Boundary of the multipart message.
multipart_form_part_limit: Limit of the number of parts allowed.
type_decoders: A sequence of type decoders to use.
Expand All @@ -113,51 +68,55 @@

fields: defaultdict[str, list[Any]] = defaultdict(list)

for form_part in parse_body(body=body, boundary=boundary, multipart_form_part_limit=multipart_form_part_limit):
file_name = None
content_type = "text/plain"
content_charset = "utf-8"
field_name = None
line_index = 2
line_end_index = 0
headers: list[tuple[str, str]] = []

while line_end_index != -1:
line_end_index = form_part.find(b"\r\n", line_index)
form_line = form_part[line_index:line_end_index].decode("utf-8")

if not form_line:
break

line_index = line_end_index + 2
colon_index = form_line.index(":")
current_idx = colon_index + 2
form_header_field = form_line[:colon_index].lower()
form_header_value, form_parameters = parse_content_header(form_line[current_idx:])

if form_header_field == "content-disposition":
field_name = form_parameters.get("name")
file_name = form_parameters.get("filename")

if file_name is None and (filename_with_asterisk := form_parameters.get("filename*")):
encoding, _, value = decode_rfc2231(filename_with_asterisk)
file_name = unquote(value, encoding=encoding or content_charset)

elif form_header_field == "content-type":
content_type = form_header_value
content_charset = form_parameters.get("charset", "utf-8")
headers.append((form_header_field, form_header_value))

if field_name:
post_data = form_part[line_index:-4].lstrip(b"\r\n")
if file_name:
form_file = UploadFile(
content_type=content_type, filename=file_name, file_data=post_data, headers=dict(headers)
)
fields[field_name].append(form_file)
elif post_data:
fields[field_name].append(post_data.decode(content_charset))
else:
fields[field_name].append(None)
chunk = await async_next(stream, b"")
if not chunk:
return fields

try:
with PushMultipartParser(boundary, max_segment_count=multipart_form_part_limit) as parser:
segment: MultipartSegment | None = None
data: UploadFile | bytearray = bytearray()
while not parser.closed:
for form_part in parser.parse(chunk):
if isinstance(form_part, MultipartSegment):
segment = form_part
if segment.filename:
data = UploadFile(
content_type=segment.content_type or "text/plain",
filename=segment.filename,
headers=dict(segment.headerlist),
)
elif form_part:
if isinstance(data, UploadFile):
await data.write(form_part)
else:
data.extend(form_part)
else:
# end of part
if segment is None:
provinzkraut marked this conversation as resolved.
Show resolved Hide resolved
# we have reached the end of a segment before we have
# received a complete header segment
raise ClientException("Unexpected eof in multipart/form-data")

Check warning on line 99 in litestar/_multipart.py

View check run for this annotation

Codecov / codecov/patch

litestar/_multipart.py#L99

Added line #L99 was not covered by tests

if isinstance(data, UploadFile):
await data.seek(0)
fields[segment.name].append(data)
elif data:
fields[segment.name].append(data.decode(segment.charset or "utf-8"))
else:
fields[segment.name].append(None)

# reset for next part
data = bytearray()
segment = None

chunk = await async_next(stream, b"")

except ParserError as exc:
raise ClientException("Invalid multipart/form-data") from exc
except ParserLimitReached:
# FIXME (3.0): This should raise a '413 - Request Entity Too Large', but for
# backwards compatibility, we keep it as a 400 for now
raise ClientException("Request Entity Too Large") from None

return {k: v if len(v) > 1 else v[0] for k, v in fields.items()}
6 changes: 3 additions & 3 deletions litestar/connection/request.py
Original file line number Diff line number Diff line change
Expand Up @@ -77,7 +77,7 @@ def __init__(self, scope: Scope, receive: Receive = empty_receive, send: Send =
"""
super().__init__(scope, receive, send)
self.is_connected: bool = True
self._body: bytes | EmptyType = Empty
self._body: bytes | EmptyType = self._connection_state.body
self._form: FormMultiDict | EmptyType = Empty
self._json: Any = Empty
self._msgpack: Any = Empty
Expand Down Expand Up @@ -264,8 +264,8 @@ async def form(self) -> FormMultiDict:
if (form_data := self._connection_state.form) is Empty:
content_type, options = self.content_type
if content_type == RequestEncodingType.MULTI_PART:
form_data = parse_multipart_form(
body=await self.body(),
form_data = await parse_multipart_form(
stream=self.stream(),
boundary=options.get("boundary", "").encode(),
multipart_form_part_limit=self.app.multipart_form_part_limit,
)
Expand Down
2 changes: 1 addition & 1 deletion litestar/datastructures/upload_file.py
Original file line number Diff line number Diff line change
Expand Up @@ -49,7 +49,7 @@ def rolled_to_disk(self) -> bool:
"""
return getattr(self.file, "_rolled", False)

async def write(self, data: bytes) -> int:
async def write(self, data: bytes | bytearray) -> int:
"""Proxy for data writing.

Args:
Expand Down
1 change: 1 addition & 0 deletions pyproject.toml
Original file line number Diff line number Diff line change
Expand Up @@ -44,6 +44,7 @@ dependencies = [
"click",
"rich>=13.0.0",
"rich-click",
"multipart>=1.2.0",
# default litestar plugins
"litestar-htmx>=0.3.0"
]
Expand Down
27 changes: 24 additions & 3 deletions tests/unit/test_kwargs/test_multipart_data.py
Original file line number Diff line number Diff line change
Expand Up @@ -186,7 +186,11 @@ def test_multipart_request_multiple_files_with_headers(tmpdir: Any) -> None:
"filename": "test2.txt",
"content": "<file2 content>",
"content_type": "text/plain",
"headers": [["content-disposition", "form-data"], ["x-custom", "f2"], ["content-type", "text/plain"]],
"headers": [
["content-disposition", 'form-data; name="test2"; filename="test2.txt"'],
["x-custom", "f2"],
["content-type", "text/plain"],
],
},
}

Expand Down Expand Up @@ -292,6 +296,7 @@ def test_multipart_request_without_charset_for_filename() -> None:
}


@pytest.mark.xfail(reason="filename* is deprecated and should not be used according to RFC-7578")
def test_multipart_request_with_asterisks_filename() -> None:
provinzkraut marked this conversation as resolved.
Show resolved Hide resolved
with create_test_client(form_handler) as client:
response = client.post(
Expand Down Expand Up @@ -456,13 +461,14 @@ async def hello_world(data: Optional[UploadFile] = Body(media_type=RequestEncodi
@pytest.mark.parametrize("limit", (1000, 100, 10))
def test_multipart_form_part_limit(limit: int) -> None:
@post("/", signature_types=[UploadFile])
async def hello_world(data: List[UploadFile] = Body(media_type=RequestEncodingType.MULTI_PART)) -> None:
assert len(data) == limit
async def hello_world(data: List[UploadFile] = Body(media_type=RequestEncodingType.MULTI_PART)) -> dict:
return {"limit": len(data)}

with create_test_client(route_handlers=[hello_world], multipart_form_part_limit=limit) as client:
data = {str(i): "a" for i in range(limit)}
response = client.post("/", files=data)
assert response.status_code == HTTP_201_CREATED
assert response.json() == {"limit": limit}

data = {str(i): "a" for i in range(limit)}
data[str(limit + 1)] = "b"
Expand Down Expand Up @@ -577,3 +583,18 @@ async def form_(request: Request, data: Annotated[AddProductFormMsgspec, Body(me
headers={"Content-Type": "multipart/form-data; boundary=1f35df74046888ceaa62d8a534a076dd"},
)
assert response.status_code == HTTP_201_CREATED


def test_invalid_multipart_raises_client_error() -> None:
with create_test_client(form_handler) as client:
response = client.post(
"/form",
content=(
b"--20b303e711c4ab8c443184ac833ab00f\r\n"
b"Content-Disposition: form-data; "
b'name="value"\r\n\r\n'
b"--20b303e711c4ab8c44318833ab00f--\r\n"
),
headers={"Content-Type": "multipart/form-data; charset=utf-8; boundary=20b303e711c4ab8c443184ac833ab00f"},
)
assert response.status_code == HTTP_400_BAD_REQUEST
11 changes: 11 additions & 0 deletions uv.lock

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

Loading