This patch addresses the problem mentioned in the "process crash

when a plpython function returns unicode" thread:

http://archives.postgresql.org/pgsql-bugs/2005-06/msg00105.php

In several places PL/Python was calling PyObject_Str() and then
PyString_AsString() without checking if the former had returned
NULL to indicate an error.  PyString_AsString() doesn't expect a
NULL argument, so passing one causes a segmentation fault.  This
patch adds checks for NULL and raises errors via PLy_elog(), which
prints details of the underlying Python exception.  The patch also
adds regression tests for these checks.  All tests pass on my
Solaris 9 box running HEAD and Python 2.4.1.

In one place the patch doesn't call PLy_elog() because that could
cause infinite recursion; see the comment I added.  I'm not sure
how to test that particular case or whether it's even possible to
get an error there: the value that the code should check is the
Python exception type, so I wonder if a NULL value "shouldn't
happen."  This patch converts NULL to "Unknown Exception" but I
wonder if an Assert() would be appropriate.

The patch is against HEAD but the same changes should be applied
to earlier versions because they have the same problem.  The patch
might not apply cleanly against earlier versions -- will the committer
take care of little differences or should I submit different versions
of the patch?

Michael Fuhr

1 files changed, 18 insertions, 3 deletions

diff --git a/src/pl/plpython/plpython.c b/src/pl/plpython/plpython.c
index 3d7230ffca9..b1d47c77820 100644
--- a/src/pl/plpython/plpython.c
+++ b/src/pl/plpython/plpython.c
@@ -29,7 +29,7 @@
  * MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS.
  *
  * IDENTIFICATION
- *	$PostgreSQL: pgsql/src/pl/plpython/plpython.c,v 1.64 2005/07/05 18:15:51 momjian Exp $
+ *	$PostgreSQL: pgsql/src/pl/plpython/plpython.c,v 1.65 2005/07/10 04:56:55 momjian Exp $
  *
  *********************************************************************
  */
@@ -517,6 +517,8 @@ PLy_modify_tuple(PLyProcedure * proc, PyObject * pltd, TriggerData *tdata,
 			if (plval != Py_None && !tupdesc->attrs[atti]->attisdropped)
 			{
 				plstr = PyObject_Str(plval);
+				if (!plstr)
+					PLy_elog(ERROR, "function \"%s\" could not modify tuple", proc->proname);
 				src = PyString_AsString(plstr);
 
 				modvalues[i] = FunctionCall3(&proc->result.out.r.atts[atti].typfunc,
@@ -774,6 +776,8 @@ PLy_function_handler(FunctionCallInfo fcinfo, PLyProcedure * proc)
 		{
 			fcinfo->isnull = false;
 			plrv_so = PyObject_Str(plrv);
+			if (!plrv_so)
+				PLy_elog(ERROR, "function \"%s\" could not create return value", proc->proname);
 			plrv_sc = PyString_AsString(plrv_so);
 			rv = FunctionCall3(&proc->result.out.d.typfunc,
 							   PointerGetDatum(plrv_sc),
@@ -2019,7 +2023,9 @@ PLy_spi_execute_plan(PyObject * ob, PyObject * list, long limit)
 		char	   *sv;
 
 		PyObject   *so = PyObject_Str(list);
-
+		if (!so)
+			PLy_elog(ERROR, "function \"%s\" could not execute plan",
+					 PLy_procedure_name(PLy_curr_procedure));
 		sv = PyString_AsString(so);
 		PLy_exception_set(PLy_exc_spi_error,
 						  "Expected sequence of %d arguments, got %d. %s",
@@ -2044,6 +2050,9 @@ PLy_spi_execute_plan(PyObject * ob, PyObject * list, long limit)
 			if (elem != Py_None)
 			{
 				so = PyObject_Str(elem);
+				if (!so)
+					PLy_elog(ERROR, "function \"%s\" could not execute plan",
+							 PLy_procedure_name(PLy_curr_procedure));
 				sv = PyString_AsString(so);
 
 				/*
@@ -2531,7 +2540,13 @@ PLy_traceback(int *xlevel)
 	else
 		vstr = "Unknown";
 
-	estr = PyString_AsString(eob);
+	/*
+	 * I'm not sure what to do if eob is NULL here -- we can't call
+	 * PLy_elog because that function calls us, so we could end up
+	 * with infinite recursion.  I'm not even sure if eob could be
+	 * NULL here -- would an Assert() be more appropriate?
+	 */
+	estr = eob ? PyString_AsString(eob) : "Unknown Exception";
 	xstr = PLy_printf("%s: %s", estr, vstr);
 
 	Py_DECREF(eob);