1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
|
From 4c4dec318f21e2eae9c69cd5b83cb54b4f9bb746 Mon Sep 17 00:00:00 2001
From: Sam Bull <git@sambull.org>
Date: Wed, 9 Jul 2025 19:55:22 +0100
Subject: [PATCH] Add trailer parsing logic (#11269) (#11287)
(cherry picked from commit 7dd4b5535e6bf9c2d2f05fde638517bff065ba74)
CVE: CVE-2025-53643
Upstream-Status: Backport
[https://github.com/aio-libs/aiohttp/commit/e8d774f635dc6d1cd3174d0e38891da5de0e2b6a]
Signed-off-by: Jiaying Song <jiaying.song.cn@windriver.com>
---
aiohttp/http_parser.py | 70 ++++++++++++++++++++++--------------------
aiohttp/multipart.py | 2 +-
2 files changed, 38 insertions(+), 34 deletions(-)
diff --git a/aiohttp/http_parser.py b/aiohttp/http_parser.py
index 1ee126940..175eb7f68 100644
--- a/aiohttp/http_parser.py
+++ b/aiohttp/http_parser.py
@@ -147,8 +147,8 @@ class HeadersParser:
headers: CIMultiDict[str] = CIMultiDict()
raw_headers = []
- lines_idx = 1
- line = lines[1]
+ lines_idx = 0
+ line = lines[lines_idx]
line_count = len(lines)
while line:
@@ -386,6 +386,7 @@ class HttpParser(abc.ABC, Generic[_MsgT]):
response_with_body=self.response_with_body,
auto_decompress=self._auto_decompress,
lax=self.lax,
+ headers_parser=self._headers_parser,
)
if not payload_parser.done:
self._payload_parser = payload_parser
@@ -405,6 +406,7 @@ class HttpParser(abc.ABC, Generic[_MsgT]):
readall=True,
auto_decompress=self._auto_decompress,
lax=self.lax,
+ headers_parser=self._headers_parser,
)
else:
if (
@@ -429,6 +431,7 @@ class HttpParser(abc.ABC, Generic[_MsgT]):
response_with_body=self.response_with_body,
auto_decompress=self._auto_decompress,
lax=self.lax,
+ headers_parser=self._headers_parser,
)
if not payload_parser.done:
self._payload_parser = payload_parser
@@ -462,6 +465,10 @@ class HttpParser(abc.ABC, Generic[_MsgT]):
eof = True
data = b""
+ if isinstance(
+ underlying_exc, (InvalidHeader, TransferEncodingError)
+ ):
+ raise
if eof:
start_pos = 0
@@ -617,7 +624,7 @@ class HttpRequestParser(HttpParser[RawRequestMessage]):
compression,
upgrade,
chunked,
- ) = self.parse_headers(lines)
+ ) = self.parse_headers(lines[1:])
if close is None: # then the headers weren't set in the request
if version_o <= HttpVersion10: # HTTP 1.0 must asks to not close
@@ -696,7 +703,7 @@ class HttpResponseParser(HttpParser[RawResponseMessage]):
compression,
upgrade,
chunked,
- ) = self.parse_headers(lines)
+ ) = self.parse_headers(lines[1:])
if close is None:
close = version_o <= HttpVersion10
@@ -727,6 +734,8 @@ class HttpPayloadParser:
response_with_body: bool = True,
auto_decompress: bool = True,
lax: bool = False,
+ *,
+ headers_parser: HeadersParser,
) -> None:
self._length = 0
self._type = ParseState.PARSE_NONE
@@ -735,6 +744,8 @@ class HttpPayloadParser:
self._chunk_tail = b""
self._auto_decompress = auto_decompress
self._lax = lax
+ self._headers_parser = headers_parser
+ self._trailer_lines: list[bytes] = []
self.done = False
# payload decompression wrapper
@@ -822,7 +833,7 @@ class HttpPayloadParser:
size_b = chunk[:i] # strip chunk-extensions
# Verify no LF in the chunk-extension
if b"\n" in (ext := chunk[i:pos]):
- exc = BadHttpMessage(
+ exc = TransferEncodingError(
f"Unexpected LF in chunk-extension: {ext!r}"
)
set_exception(self.payload, exc)
@@ -843,7 +854,7 @@ class HttpPayloadParser:
chunk = chunk[pos + len(SEP) :]
if size == 0: # eof marker
- self._chunk = ChunkState.PARSE_MAYBE_TRAILERS
+ self._chunk = ChunkState.PARSE_TRAILERS
if self._lax and chunk.startswith(b"\r"):
chunk = chunk[1:]
else:
@@ -881,38 +892,31 @@ class HttpPayloadParser:
self._chunk_tail = chunk
return False, b""
- # if stream does not contain trailer, after 0\r\n
- # we should get another \r\n otherwise
- # trailers needs to be skiped until \r\n\r\n
- if self._chunk == ChunkState.PARSE_MAYBE_TRAILERS:
- head = chunk[: len(SEP)]
- if head == SEP:
- # end of stream
- self.payload.feed_eof()
- return True, chunk[len(SEP) :]
- # Both CR and LF, or only LF may not be received yet. It is
- # expected that CRLF or LF will be shown at the very first
- # byte next time, otherwise trailers should come. The last
- # CRLF which marks the end of response might not be
- # contained in the same TCP segment which delivered the
- # size indicator.
- if not head:
- return False, b""
- if head == SEP[:1]:
- self._chunk_tail = head
- return False, b""
- self._chunk = ChunkState.PARSE_TRAILERS
-
- # read and discard trailer up to the CRLF terminator
if self._chunk == ChunkState.PARSE_TRAILERS:
pos = chunk.find(SEP)
- if pos >= 0:
- chunk = chunk[pos + len(SEP) :]
- self._chunk = ChunkState.PARSE_MAYBE_TRAILERS
- else:
+ if pos < 0: # No line found
self._chunk_tail = chunk
return False, b""
+ line = chunk[:pos]
+ chunk = chunk[pos + len(SEP) :]
+ if SEP == b"\n": # For lax response parsing
+ line = line.rstrip(b"\r")
+ self._trailer_lines.append(line)
+
+ # \r\n\r\n found, end of stream
+ if self._trailer_lines[-1] == b"":
+ # Headers and trailers are defined the same way,
+ # so we reuse the HeadersParser here.
+ try:
+ trailers, raw_trailers = self._headers_parser.parse_headers(
+ self._trailer_lines
+ )
+ finally:
+ self._trailer_lines.clear()
+ self.payload.feed_eof()
+ return True, chunk
+
# Read all bytes until eof
elif self._type == ParseState.PARSE_UNTIL_EOF:
self.payload.feed_data(chunk, len(chunk))
diff --git a/aiohttp/multipart.py b/aiohttp/multipart.py
index 9cd49bb5d..2457f80ec 100644
--- a/aiohttp/multipart.py
+++ b/aiohttp/multipart.py
@@ -718,7 +718,7 @@ class MultipartReader:
raise ValueError(f"Invalid boundary {chunk!r}, expected {self._boundary!r}")
async def _read_headers(self) -> "CIMultiDictProxy[str]":
- lines = [b""]
+ lines = []
while True:
chunk = await self._content.readline()
chunk = chunk.strip()
--
2.34.1
|
