From: Bruce Momjian <pgman@candle.pha.pa.us>
Message-Id: <200303210555.h2L5tKG10364@candle.pha.pa.us>
Subject: Re: I want to send comments to the backend!
In-Reply-To: <10177.1048225294@sss.pgh.pa.us>
To: Tom Lane <tgl@sss.pgh.pa.us>
Date: Fri, 21 Mar 2003 00:55:20 -0500 (EST)
Cc: =?ISO-8859-1?Q?Gerhard_H=E4ring?= <lists@ghaering.de>,
	pgsql-interfaces@postgresql.org
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary=ELM1048226120-11733-3_
Content-Transfer-Encoding: 7bit

--ELM1048226120-11733-3_
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=US-ASCII

Tom Lane wrote:
> Bruce Momjian <pgman@candle.pha.pa.us> writes:
> > OK, this patch fixes the comment passing bug.  I remove the strspn() and
> > added a boolean to test if any parsetree had been generated --- if not,
> > I call NullCommand().
> 
> Seems like the hard way.  I had in mind a quick
> 
> 	if (parsetree_list == NIL)
> 	{
> 		NullCommand(dest);
> 		return;
> 	}
> 
> before entering the main loop.  It'd take a small amount of reordering
> of the existing code to make this happen without adding any more code
> than that, but it looked doable.

The problem is that there is so much startup/shutdown in that function
that adding a quick exit point seems too error-prone for maintenance,
rather than keeping the existing structure.

I used your idea of testing just 'parsetree_list' and the patch is now
smaller.

-- 
  Bruce Momjian                        |  http://candle.pha.pa.us
  pgman@candle.pha.pa.us               |  (610) 359-1001
  +  If your life is a hard drive,     |  13 Roberts Road
  +  Christ can be your backup.        |  Newtown Square, Pennsylvania 19073

--ELM1048226120-11733-3_
Content-Transfer-Encoding: 7bit
Content-Type: text/plain
Content-Disposition: inline; filename="/pgpatches/empty"

Index: src/backend/tcop/postgres.c
===================================================================
RCS file: /cvsroot/pgsql-server/src/backend/tcop/postgres.c,v
retrieving revision 1.318
diff -c -c -r1.318 postgres.c
*** src/backend/tcop/postgres.c	20 Mar 2003 07:02:10 -0000	1.318
--- src/backend/tcop/postgres.c	21 Mar 2003 05:52:56 -0000
***************
*** 927,932 ****
--- 927,936 ----
  		EndCommand(commandTag, dest);
  	}							/* end loop over parsetrees */
  
+ 	/* No parsetree - return empty result */
+ 	if (!parsetree_list && IsUnderPostmaster)
+ 		NullCommand(dest);
+ 
  	/*
  	 * Close down transaction statement, if one is open. (Note that this
  	 * will only happen if the querystring was empty.)
***************
*** 1995,2030 ****
  				 * 'Q' indicates a user query
  				 */
  			case 'Q':
! 				if (strspn(parser_input->data, " \t\r\n") == parser_input->len)
! 				{
! 					/*
! 					 * if there is nothing in the input buffer, don't
! 					 * bother trying to parse and execute anything; just
! 					 * send back a quick NullCommand response.
! 					 */
! 					if (IsUnderPostmaster)
! 						NullCommand(Remote);
! 				}
! 				else
! 				{
! 					/*
! 					 * otherwise, process the input string.
! 					 *
! 					 * Note: transaction command start/end is now done within
! 					 * pg_exec_query_string(), not here.
! 					 */
! 					if (log_statement_stats)
! 						ResetUsage();
  
! 					pgstat_report_activity(parser_input->data);
  
! 					pg_exec_query_string(parser_input,
! 										 whereToSendOutput,
! 										 QueryContext);
  
! 					if (log_statement_stats)
! 						ShowUsage("QUERY STATISTICS");
! 				}
  				break;
  
  				/*
--- 1999,2021 ----
  				 * 'Q' indicates a user query
  				 */
  			case 'Q':
! 				/*
! 				 * otherwise, process the input string.
! 				 *
! 				 * Note: transaction command start/end is now done within
! 				 * pg_exec_query_string(), not here.
! 				 */
! 				if (log_statement_stats)
! 					ResetUsage();
  
! 				pgstat_report_activity(parser_input->data);
  
! 				pg_exec_query_string(parser_input,
! 									 whereToSendOutput,
! 									 QueryContext);
  
! 				if (log_statement_stats)
! 					ShowUsage("QUERY STATISTICS");
  				break;
  
  				/*

--ELM1048226120-11733-3_--