pythongh-127081: use re-entrant variants of get{proto,serv}by{name,port}

Add configure tests and defines for getservbyname_r, getservbyport_r, and
getprotobyname_r. Use these if available, otherwise fallback to the
thread-unsafe variants.

Add a unit test to exercise getprotobyname, which is currently untested.

TODO:
 - Are there any platforms which define the unsafe variants but not the
   re-entrant ones? If not we can simplify the #ifdef hell somewhat.
 - Do the re-entrant functions have the same signature on all platforms?
 - These changes follow the existing code's practice: allocate a fixed-size
   (and overly large) buffer, and don't properly handle the error case if it is
   too small. Should this be fixed? If so should existing code also be fixed?

Author: duaneg

Lib/test/test_socket.py

@@ -1297,6 +1297,10 @@ def testGetServBy(self):
         self.assertRaises(OverflowError, socket.getservbyport, -1)
         self.assertRaises(OverflowError, socket.getservbyport, 65536)
 
+    def testGetProtoByName(self):
+        self.assertEqual(socket.getprotobyname('tcp'), 6)
+        self.assertRaises(OSError, socket.getprotobyname, 'non-existent proto')
+
     def testDefaultTimeout(self):
         # Testing default timeout
         # The default timeout should initially be None

Misc/NEWS.d/next/Library/2025-04-17-00-14-41.gh-issue-127081.aEM3Hk.rst

@@ -0,0 +1,2 @@
+Fix thread safety issues with getservbyname, getservbyport, and
+getprotobyname.

Modules/socketmodule.c

@@ -6269,7 +6269,7 @@ Return the true host name, a list of aliases, and a list of IP addresses,\n\
 for a host.  The host argument is a string giving a host name or IP number.");
 #endif
 
-#ifdef HAVE_GETSERVBYNAME
+#if defined(HAVE_GETSERVBYNAME_R) || defined (HAVE_GETSERVBYNAME)
 /* Python interface to getservbyname(name).
    This only returns the port number, since the other info is already
    known or not useful (like the list of aliases). */
@@ -6279,6 +6279,12 @@ static PyObject *
 socket_getservbyname(PyObject *self, PyObject *args)
 {
     const char *name, *proto=NULL;
+#ifdef HAVE_GETSERVBYNAME_R
+    struct servent entry;
+    /* TODO: The man page says 1024 is usually enough, start with that and
+       retry if insufficient? */
+    char buf[16384];
+#endif
     struct servent *sp;
     if (!PyArg_ParseTuple(args, "s|s:getservbyname", &name, &proto))
         return NULL;
@@ -6288,7 +6294,11 @@ socket_getservbyname(PyObject *self, PyObject *args)
     }
 
     Py_BEGIN_ALLOW_THREADS
+#ifdef HAVE_GETSERVBYNAME_R
+    getservbyname_r(name, proto, &entry, buf, sizeof(buf), &sp);
+#else
     sp = getservbyname(name, proto);
+#endif
     Py_END_ALLOW_THREADS
     if (sp == NULL) {
         PyErr_SetString(PyExc_OSError, "service/proto not found");
@@ -6305,7 +6315,7 @@ The optional protocol name, if given, should be 'tcp' or 'udp',\n\
 otherwise any protocol will match.");
 #endif
 
-#ifdef HAVE_GETSERVBYPORT
+#if defined(HAVE_GETSERVBYPORT_R) || defined (HAVE_GETSERVBYPORT)
 /* Python interface to getservbyport(port).
    This only returns the service name, since the other info is already
    known or not useful (like the list of aliases). */
@@ -6316,6 +6326,12 @@ socket_getservbyport(PyObject *self, PyObject *args)
 {
     int port;
     const char *proto=NULL;
+#ifdef HAVE_GETSERVBYPORT_R
+    struct servent entry;
+    /* TODO: The man page says 1024 is usually enough, start with that and
+       retry if insufficient? */
+    char buf[16384];
+#endif
     struct servent *sp;
     if (!PyArg_ParseTuple(args, "i|s:getservbyport", &port, &proto))
         return NULL;
@@ -6331,7 +6347,11 @@ socket_getservbyport(PyObject *self, PyObject *args)
     }
 
     Py_BEGIN_ALLOW_THREADS
+#ifdef HAVE_GETSERVBYPORT_R
+    getservbyport_r(htons((short)port), proto, &entry, buf, sizeof(buf), &sp);
+#else
     sp = getservbyport(htons((short)port), proto);
+#endif
     Py_END_ALLOW_THREADS
     if (sp == NULL) {
         PyErr_SetString(PyExc_OSError, "port/proto not found");
@@ -6348,7 +6368,7 @@ The optional protocol name, if given, should be 'tcp' or 'udp',\n\
 otherwise any protocol will match.");
 #endif
 
-#ifdef HAVE_GETPROTOBYNAME
+#if defined(HAVE_GETPROTOBYNAME_R) || defined (HAVE_GETPROTOBYNAME)
 /* Python interface to getprotobyname(name).
    This only returns the protocol number, since the other info is
    already known or not useful (like the list of aliases). */
@@ -6358,11 +6378,21 @@ static PyObject *
 socket_getprotobyname(PyObject *self, PyObject *args)
 {
     const char *name;
+#ifdef HAVE_GETPROTOBYNAME_R
+    struct protoent entry;
+    /* TODO: The man page says 1024 is usually enough, start with that and
+       retry if insufficient? */
+    char buf[16384];
+#endif
     struct protoent *sp;
     if (!PyArg_ParseTuple(args, "s:getprotobyname", &name))
         return NULL;
     Py_BEGIN_ALLOW_THREADS
+#ifdef HAVE_GETPROTOBYNAME_R
+    getprotobyname_r(name, &entry, buf, sizeof(buf), &sp);
+#else
     sp = getprotobyname(name);
+#endif
     Py_END_ALLOW_THREADS
     if (sp == NULL) {
         PyErr_SetString(PyExc_OSError, "protocol not found");
@@ -7415,15 +7445,15 @@ static PyMethodDef socket_methods[] = {
     {"sethostname",             socket_sethostname,
      METH_VARARGS,  sethostname_doc},
 #endif
-#ifdef HAVE_GETSERVBYNAME
+#if defined(HAVE_GETSERVBYNAME_R) || defined (HAVE_GETSERVBYNAME)
     {"getservbyname",           socket_getservbyname,
      METH_VARARGS, getservbyname_doc},
 #endif
-#ifdef HAVE_GETSERVBYPORT
+#if defined(HAVE_GETSERVBYPORT_R) || defined (HAVE_GETSERVBYPORT)
     {"getservbyport",           socket_getservbyport,
      METH_VARARGS, getservbyport_doc},
 #endif
-#ifdef HAVE_GETPROTOBYNAME
+#if defined (HAVE_GETPROTOBYNAME_R) || defined (HAVE_GETPROTOBYNAME)
     {"getprotobyname",          socket_getprotobyname,
      METH_VARARGS, getprotobyname_doc},
 #endif

configure

@@ -5395,6 +5395,9 @@ PY_CHECK_NETDB_FUNC([getservbyport])
 PY_CHECK_NETDB_FUNC([gethostbyname])
 PY_CHECK_NETDB_FUNC([gethostbyaddr])
 PY_CHECK_NETDB_FUNC([getprotobyname])
+PY_CHECK_NETDB_FUNC([getservbyname_r])
+PY_CHECK_NETDB_FUNC([getservbyport_r])
+PY_CHECK_NETDB_FUNC([getprotobyname_r])
 
 dnl PY_CHECK_SOCKET_FUNC(FUNCTION)
 AC_DEFUN([PY_CHECK_SOCKET_FUNC], [PY_CHECK_FUNC([$1], [

configure.ac

@@ -557,6 +557,9 @@
 /* Define if you have the 'getprotobyname' function. */
 #undef HAVE_GETPROTOBYNAME
 
+/* Define if you have the 'getprotobyname_r' function. */
+#undef HAVE_GETPROTOBYNAME_R
+
 /* Define to 1 if you have the 'getpwent' function. */
 #undef HAVE_GETPWENT
 
@@ -587,9 +590,15 @@
 /* Define if you have the 'getservbyname' function. */
 #undef HAVE_GETSERVBYNAME
 
+/* Define if you have the 'getservbyname_r' function. */
+#undef HAVE_GETSERVBYNAME_R
+
 /* Define if you have the 'getservbyport' function. */
 #undef HAVE_GETSERVBYPORT
 
+/* Define if you have the 'getservbyport_r' function. */
+#undef HAVE_GETSERVBYPORT_R
+
 /* Define to 1 if you have the 'getsid' function. */
 #undef HAVE_GETSID