Re: [HACKERS] Allowing same cursor name in nested levels

2011-08-17 Thread Jeevan Chalke
On Tue, Aug 16, 2011 at 7:25 PM, Tom Lane t...@sss.pgh.pa.us wrote:

 Jeevan Chalke jeevan.cha...@enterprisedb.com writes:
  1. I have two functions func1 and func2.
  2. func1 calls func2
  3. Both has cursor with same name, say mycursor
  4. Somehow I forgot closing it
  5. executing func1 throws an error 'cursor mycursor already in use'

  Is this expected behavior???

 Yes ... or at least, it's always been like that.

  I just mingled around the code and later appended a cursor count to the
  cursor name to allow same cursor name in nested levels.

 That would break code that expects the cursor name to be what it said
 it should be.  It is documented that you can refer to cursors by name
 across multiple functions.


Hmm... got it.

Thanks for the clarification.



regards, tom lane




-- 
Jeevan B Chalke
Senior Software Engineer, RD
EnterpriseDB Corporation
The Enterprise PostgreSQL Company

Phone: +91 20 30589500

Website: www.enterprisedb.com
EnterpriseDB Blog: http://blogs.enterprisedb.com/
Follow us on Twitter: http://www.twitter.com/enterprisedb

This e-mail message (and any attachment) is intended for the use of the
individual or entity to whom it is addressed. This message contains
information from EnterpriseDB Corporation that may be privileged,
confidential, or exempt from disclosure under applicable law. If you are not
the intended recipient or authorized to receive this for the intended
recipient, any use, dissemination, distribution, retention, archiving, or
copying of this communication is strictly prohibited. If you have received
this e-mail in error, please notify the sender immediately by reply e-mail
and delete this message.


[HACKERS] Allowing same cursor name in nested levels

2011-08-16 Thread Jeevan Chalke
Hi Tom,

While going through few test-cases, I found that we cannot have two opened
cursors with same name even though they are in two different functions. Here
is what I mean:

1. I have two functions func1 and func2.
2. func1 calls func2
3. Both has cursor with same name, say mycursor
4. Somehow I forgot closing it
5. executing func1 throws an error 'cursor mycursor already in use'


Is this expected behavior???

I just mingled around the code and later appended a cursor count to the
cursor name to allow same cursor name in nested levels. I have run make
check and didn't find any side-effect other than one expected output change.

PFA, patch for the same.

Please let me know your views / suggestions.

Thanks

Test-case:
===

CREATE OR REPLACE FUNCTION func1() RETURNS void AS $$
DECLARE
 res int;
 mycursor CURSOR FOR SELECT 'foo';
BEGIN
 OPEN mycursor;
 SELECT func2() INTO res;
END;
$$ LANGUAGE plpgsql;

CREATE OR REPLACE FUNCTION func2() RETURNS int AS $$
DECLARE
 mycursor CURSOR FOR SELECT 'bar';
BEGIN
 OPEN mycursor;
 return 1;
END;
$$ LANGUAGE plpgsql;

SELECT func1();
ERROR:  cursor mycursor already in use
CONTEXT:  PL/pgSQL function func2 line 5 at OPEN
SQL statement SELECT func2()
PL/pgSQL function func1 line 7 at SQL statement




-- 
Jeevan B Chalke
Senior Software Engineer, RD
EnterpriseDB Corporation
The Enterprise PostgreSQL Company

Phone: +91 20 30589500

Website: www.enterprisedb.com
EnterpriseDB Blog: http://blogs.enterprisedb.com/
Follow us on Twitter: http://www.twitter.com/enterprisedb

This e-mail message (and any attachment) is intended for the use of the
individual or entity to whom it is addressed. This message contains
information from EnterpriseDB Corporation that may be privileged,
confidential, or exempt from disclosure under applicable law. If you are not
the intended recipient or authorized to receive this for the intended
recipient, any use, dissemination, distribution, retention, archiving, or
copying of this communication is strictly prohibited. If you have received
this e-mail in error, please notify the sender immediately by reply e-mail
and delete this message.
diff --git a/src/pl/plpgsql/src/gram.y b/src/pl/plpgsql/src/gram.y
index 92b54dd..46e7442 100644
--- a/src/pl/plpgsql/src/gram.y
+++ b/src/pl/plpgsql/src/gram.y
@@ -509,6 +509,7 @@ decl_statement	: decl_varname decl_const decl_datatype decl_collate decl_notnull
 		char		buf[1024];
 		char		*cp1;
 		char		*cp2;
+		static unsigned int cursor_count = 0;
 
 		/* pop local namespace for cursor args */
 		plpgsql_ns_pop();
@@ -539,7 +540,13 @@ decl_statement	: decl_varname decl_const decl_datatype decl_collate decl_notnull
 *cp2++ = *cp1;
 			*cp2++ = *cp1++;
 		}
-		strcpy(cp2, '::pg_catalog.refcursor);
+
+		/*
+		 * Append cursor count after cursor name to remove
+		 * conflict of same named cursor in different
+		 * nested levels.
+	 	 */
+		snprintf(cp2, sizeof(buf) - strlen(buf), :%u'::pg_catalog.refcursor, cursor_count++);
 		curname_def-query = pstrdup(buf);
 		new-default_val = curname_def;
 
diff --git a/src/test/regress/expected/plpgsql.out b/src/test/regress/expected/plpgsql.out
index bed34c8..4bdccf3 100644
--- a/src/test/regress/expected/plpgsql.out
+++ b/src/test/regress/expected/plpgsql.out
@@ -3284,15 +3284,15 @@ begin
 end;
 $$ language plpgsql;
 select forc01();
-NOTICE:  5 from c
-NOTICE:  6 from c
-NOTICE:  7 from c
-NOTICE:  9 from c
-NOTICE:  10 from c
-NOTICE:  41 from c2
-NOTICE:  42 from c2
-NOTICE:  43 from c2
-NOTICE:  after loop, c2 = c2
+NOTICE:  5 from c:5
+NOTICE:  6 from c:5
+NOTICE:  7 from c:5
+NOTICE:  9 from c:5
+NOTICE:  10 from c:5
+NOTICE:  41 from c2:6
+NOTICE:  42 from c2:6
+NOTICE:  43 from c2:6
+NOTICE:  after loop, c2 = c2:6
 NOTICE:  41 from special_name
 NOTICE:  42 from special_name
 NOTICE:  43 from special_name

-- 
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers


Re: [HACKERS] Allowing same cursor name in nested levels

2011-08-16 Thread Tom Lane
Jeevan Chalke jeevan.cha...@enterprisedb.com writes:
 1. I have two functions func1 and func2.
 2. func1 calls func2
 3. Both has cursor with same name, say mycursor
 4. Somehow I forgot closing it
 5. executing func1 throws an error 'cursor mycursor already in use'

 Is this expected behavior???

Yes ... or at least, it's always been like that.

 I just mingled around the code and later appended a cursor count to the
 cursor name to allow same cursor name in nested levels.

That would break code that expects the cursor name to be what it said
it should be.  It is documented that you can refer to cursors by name
across multiple functions.

regards, tom lane

-- 
Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-hackers