[issue16201] socket.gethostbyname incorrectly parses ip
Charles-François Natali
report at bugs.python.org
Thu Aug 29 16:13:59 CEST 2013
Charles-François Natali added the comment:
Here's a patch using inet_pton() if available, otherwise inet_aton()
if available, otherwise fallback to getaddrinfo().
This should work on every platform, but if a platform has neither
inet_pton() nor inet_aton(), calling getaddrinfo() will incur an
overhead.
All Unices have either inet_aton() or inet_pton().
For Windows >= Vista, inet_pton() is available.
I'm not sure whether XP has one of them: if not, then we might keep
the hand-parsing as a fallback.
This adds support for numeric IPv6 addresses (if inet_pton() is
available), which notably speeds up sendto() on IPv6 sockets, and
gives a tiny speedup for IPv4 sockets:
before:
$ ./python -m timeit -s "import socket; s =
socket.socket(socket.AF_INET, socket.SOCK_DGRAM); DATA = b'x'"
"s.sendto(DATA, ('127.0.0.1', 4242))"
100000 loops, best of 3: 13.2 usec per loop
$ ./python -m timeit -s "import socket; s =
socket.socket(socket.AF_INET6, socket.SOCK_DGRAM); DATA = b'x'"
"s.sendto(DATA, ('::1', 4242))"
10000 loops, best of 3: 32.1 usec per loop
after:
$ ./python -m timeit -s "import socket; s =
socket.socket(socket.AF_INET, socket.SOCK_DGRAM); DATA = b'x'"
"s.sendto(DATA, ('127.0.0.1', 4242))"
100000 loops, best of 3: 11.5 usec per loop
$ ./python -m timeit -s "import socket; s =
socket.socket(socket.AF_INET6, socket.SOCK_DGRAM); DATA = b'x'"
"s.sendto(DATA, ('::1', 4242))"
100000 loops, best of 3: 12.8 usec per loop
Note that if the IPv6 address contains a scope ID ('%' suffix), then
we fallback to getaddrinfo() which handles it properly (the scope ID
can be an interface index but also an interface name).
----------
Added file: http://bugs.python.org/file31506/parse_inet.diff
_______________________________________
Python tracker <report at bugs.python.org>
<http://bugs.python.org/issue16201>
_______________________________________
-------------- next part --------------
diff -r 4b3238923b01 Lib/test/test_socket.py
--- a/Lib/test/test_socket.py Fri May 10 19:57:44 2013 -0700
+++ b/Lib/test/test_socket.py Thu Aug 29 16:10:40 2013 +0200
@@ -761,6 +761,31 @@
if not fqhn in all_host_names:
self.fail("Error testing host resolution mechanisms. (fqdn: %s, all: %s)" % (fqhn, repr(all_host_names)))
+ def test_hosts_resolution(self):
+ valid_hosts = [support.HOST]
+ valid_hosts_ipv6 = [support.HOSTv6]
+
+ invalid_hosts = ['0.1.1.1.', 'a1.1.1.1', '10.1a.1.1',
+ '4294967306.4294967296.4294967296.1']
+ invalid_hosts_ipv6 = ['::1q', '::1::2', '1:1:1:1:1:1:1:1:1']
+
+ for host in valid_hosts:
+ self.assertEqual(socket.gethostbyname(host), host)
+
+ for host in invalid_hosts:
+ self.assertRaises(OSError, socket.gethostbyname, host)
+
+ # gethostbyaddr() support IPv6, contrarily to gethosbyname()
+ if support.IPV6_ENABLED:
+ valid_hosts.extend(valid_hosts_ipv6)
+ invalid_hosts.extend(invalid_hosts_ipv6)
+
+ for host in valid_hosts:
+ self.assertIn(host, socket.gethostbyaddr(host)[2])
+
+ for host in invalid_hosts:
+ self.assertRaises(OSError, socket.gethostbyaddr, host)
+
@unittest.skipUnless(hasattr(socket, 'sethostname'), "test needs socket.sethostname()")
@unittest.skipUnless(hasattr(socket, 'gethostname'), "test needs socket.gethostname()")
def test_sethostname(self):
diff -r 4b3238923b01 Modules/socketmodule.c
--- a/Modules/socketmodule.c Fri May 10 19:57:44 2013 -0700
+++ b/Modules/socketmodule.c Thu Aug 29 16:10:40 2013 +0200
@@ -799,8 +799,6 @@
{
struct addrinfo hints, *res;
int error;
- int d1, d2, d3, d4;
- char ch;
memset((void *) addr_ret, '\0', sizeof(*addr_ret));
if (name[0] == '\0') {
@@ -865,20 +863,49 @@
sin->sin_addr.s_addr = INADDR_BROADCAST;
return sizeof(sin->sin_addr);
}
- if (sscanf(name, "%d.%d.%d.%d%c", &d1, &d2, &d3, &d4, &ch) == 4 &&
- 0 <= d1 && d1 <= 255 && 0 <= d2 && d2 <= 255 &&
- 0 <= d3 && d3 <= 255 && 0 <= d4 && d4 <= 255) {
- struct sockaddr_in *sin;
- sin = (struct sockaddr_in *)addr_ret;
- sin->sin_addr.s_addr = htonl(
- ((long) d1 << 24) | ((long) d2 << 16) |
- ((long) d3 << 8) | ((long) d4 << 0));
- sin->sin_family = AF_INET;
+ /* avoid a name resolution in case of numeric address */
+#ifdef HAVE_INET_PTON
+ /* check for an IPv4 address */
+ if (af == AF_UNSPEC || af == AF_INET) {
+ struct sockaddr_in *sin = (struct sockaddr_in *)addr_ret;
+ if (inet_pton(AF_INET, name, &sin->sin_addr) > 0) {
+ sin->sin_family = AF_INET;
#ifdef HAVE_SOCKADDR_SA_LEN
- sin->sin_len = sizeof(*sin);
-#endif
- return 4;
- }
+ sin->sin_len = sizeof(*sin);
+#endif
+ return 4;
+ }
+ }
+#ifdef ENABLE_IPV6
+ /* check for an IPv6 address - if the address contains a scope ID, we
+ * fallback to getaddrinfo(), which can handle translation from interface
+ * name to interface index */
+ if ((af == AF_UNSPEC || af == AF_INET6) && !strchr(name, '%')) {
+ struct sockaddr_in6 *sin = (struct sockaddr_in6 *)addr_ret;
+ if (inet_pton(AF_INET6, name, &sin->sin6_addr) > 0) {
+ sin->sin6_family = AF_INET6;
+#ifdef HAVE_SOCKADDR_SA_LEN
+ sin->sin6_len = sizeof(*sin);
+#endif
+ return 16;
+ }
+ }
+#endif /* ENABLE_IPV6 */
+#elif defined(HAVE_INET_ATON)
+ /* check for an IPv4 address */
+ if (af == AF_INET || af == AF_UNSPEC) {
+ struct sockaddr_in *sin = (struct sockaddr_in *)addr_ret;
+ if (inet_aton(name, &sin->sin_addr) > 0) {
+ sin->sin_family = AF_INET;
+#ifdef HAVE_SOCKADDR_SA_LEN
+ sin->sin_len = sizeof(*sin);
+#endif
+ return 4;
+ }
+ }
+#endif /* HAVE_INET_PTON */
+
+ /* perform a name resolution */
memset(&hints, 0, sizeof(hints));
hints.ai_family = af;
Py_BEGIN_ALLOW_THREADS
More information about the Python-bugs-list
mailing list