Avoid dump/reload problems when using both plpython2 and plpython3.

Commit 80371601 installed a safeguard against loading plpython2
and plpython3 at the same time, but asserted that both could still be
used in the same database, just not in the same session.  However, that's
not actually all that practical because dumping and reloading will fail
(since both libraries necessarily get loaded into the restoring session).
pg_upgrade is even worse, because it checks for missing libraries by
loading every .so library mentioned in the entire installation into one
session, so that you can have only one across the whole cluster.

We can improve matters by not throwing the error immediately in _PG_init,
but only when and if we're asked to do something that requires calling
into libpython.  This ameliorates both of the above situations, since
while execution of CREATE LANGUAGE, CREATE FUNCTION, etc will result in
loading plpython, it isn't asked to do anything interesting (at least
not if check_function_bodies is off, as it will be during a restore).

It's possible that this opens some corner-case holes in which a crash
could be provoked with sufficient effort.  However, since plpython
only exists as an untrusted language, any such crash would require
superuser privileges, making it "don't do that" not a security issue.
To reduce the hazards in this area, the error is still FATAL when it
does get thrown.

Per a report from Paul Jones.  Back-patch to 9.2, which is as far back
as the patch applies without work.  (It could be made to work in 9.1,
but given the lack of previous complaints, I'm disinclined to expend
effort so far back.  We've been pretty desultory about support for
Python 3 in 9.1 anyway.)

src/pl/plpython/plpy_main.c

@@ -63,6 +63,9 @@ static void PLy_init_interp(void);
 static PLyExecutionContext *PLy_push_execution_context(void);
 static void PLy_pop_execution_context(void);
 
+/* static state for Python library conflict detection */
+static int *plpython_version_bitmask_ptr = NULL;
+static int	plpython_version_bitmask = 0;
 static const int plpython_python_version = PY_MAJOR_VERSION;
 
 /* initialize global variables */
@@ -75,28 +78,81 @@ static PLyExecutionContext *PLy_execution_contexts = NULL;
 void
 _PG_init(void)
 {
-	/* Be sure we do initialization only once (should be redundant now) */
-	static bool inited = false;
+	int		  **bitmask_ptr;
 	const int **version_ptr;
 
-	if (inited)
-		return;
+	/*
+	 * Set up a shared bitmask variable telling which Python version(s) are
+	 * loaded into this process's address space.  If there's more than one, we
+	 * cannot call into libpython for fear of causing crashes.  But postpone
+	 * the actual failure for later, so that operations like pg_restore can
+	 * load more than one plpython library so long as they don't try to do
+	 * anything much with the language.
+	 */
+	bitmask_ptr = (int **) find_rendezvous_variable("plpython_version_bitmask");
+	if (!(*bitmask_ptr))		/* am I the first? */
+		*bitmask_ptr = &plpython_version_bitmask;
+	/* Retain pointer to the agreed-on shared variable ... */
+	plpython_version_bitmask_ptr = *bitmask_ptr;
+	/* ... and announce my presence */
+	*plpython_version_bitmask_ptr |= (1 << PY_MAJOR_VERSION);
+
+	/*
+	 * This should be safe even in the presence of conflicting plpythons, and
+	 * it's necessary to do it here for the next error to be localized.
+	 */
+	pg_bindtextdomain(TEXTDOMAIN);
 
-	/* Be sure we don't run Python 2 and 3 in the same session (might crash) */
+	/*
+	 * We used to have a scheme whereby PL/Python would fail immediately if
+	 * loaded into a session in which a conflicting libpython is already
+	 * present.  We don't like to do that anymore, but it seems possible that
+	 * a plpython library adhering to the old convention is present in the
+	 * session, in which case we have to fail.  We detect an old library if
+	 * plpython_python_version is already defined but the indicated version
+	 * isn't reflected in plpython_version_bitmask.  Otherwise, set the
+	 * variable so that the right thing happens if an old library is loaded
+	 * later.
+	 */
 	version_ptr = (const int **) find_rendezvous_variable("plpython_python_version");
 	if (!(*version_ptr))
 		*version_ptr = &plpython_python_version;
 	else
 	{
-		if (**version_ptr != plpython_python_version)
+		if ((*plpython_version_bitmask_ptr & (1 << **version_ptr)) == 0)
 			ereport(FATAL,
 					(errmsg("Python major version mismatch in session"),
 					 errdetail("This session has previously used Python major version %d, and it is now attempting to use Python major version %d.",
 							   **version_ptr, plpython_python_version),
 					 errhint("Start a new session to use a different Python major version.")));
 	}
+}
 
-	pg_bindtextdomain(TEXTDOMAIN);
+/*
+ * Perform one-time setup of PL/Python, after checking for a conflict
+ * with other versions of Python.
+ */
+static void
+PLy_initialize(void)
+{
+	static bool inited = false;
+
+	/*
+	 * Check for multiple Python libraries before actively doing anything with
+	 * libpython.  This must be repeated on each entry to PL/Python, in case a
+	 * conflicting library got loaded since we last looked.
+	 *
+	 * It is attractive to weaken this error from FATAL to ERROR, but there
+	 * would be corner cases, so it seems best to be conservative.
+	 */
+	if (*plpython_version_bitmask_ptr != (1 << PY_MAJOR_VERSION))
+		ereport(FATAL,
+				(errmsg("multiple Python libraries are present in session"),
+				 errdetail("Only one Python major version can be used in one session.")));
+
+	/* The rest should only be done once per session */
+	if (inited)
+		return;
 
 #if PY_MAJOR_VERSION >= 3
 	PyImport_AppendInittab("plpy", PyInit_plpy);
@@ -120,7 +176,7 @@ _PG_init(void)
 }
 
 /*
- * This should only be called once from _PG_init. Initialize the Python
+ * This should be called only once, from PLy_initialize. Initialize the Python
  * interpreter and global data.
  */
 static void
@@ -155,9 +211,10 @@ plpython_validator(PG_FUNCTION_ARGS)
 		PG_RETURN_VOID();
 
 	if (!check_function_bodies)
-	{
 		PG_RETURN_VOID();
-	}
+
+	/* Do this only after making sure we need to do something */
+	PLy_initialize();
 
 	/* Get the new function's pg_proc entry */
 	tuple = SearchSysCache1(PROCOID, ObjectIdGetDatum(funcoid));
@@ -191,6 +248,8 @@ plpython_call_handler(PG_FUNCTION_ARGS)
 	PLyExecutionContext *exec_ctx;
 	ErrorContextCallback plerrcontext;
 
+	PLy_initialize();
+
 	/* Note: SPI_finish() happens in plpy_exec.c, which is dubious design */
 	if (SPI_connect() != SPI_OK_CONNECT)
 		elog(ERROR, "SPI_connect failed");
@@ -266,6 +325,8 @@ plpython_inline_handler(PG_FUNCTION_ARGS)
 	PLyExecutionContext *exec_ctx;
 	ErrorContextCallback plerrcontext;
 
+	PLy_initialize();
+
 	/* Note: SPI_finish() happens in plpy_exec.c, which is dubious design */
 	if (SPI_connect() != SPI_OK_CONNECT)
 		elog(ERROR, "SPI_connect failed");