Re: [HACKERS] BEGIN inside transaction should be an error - Mailing list pgsql-patches

From Gurjeet Singh
Subject Re: [HACKERS] BEGIN inside transaction should be an error
Date
Msg-id 65937bea0605260900q163fbc24ie9294089246adce9@mail.gmail.com
Whole thread Raw
In response to Re: [HACKERS] BEGIN inside transaction should be an error  (Alvaro Herrera <alvherre@commandprompt.com>)
Responses Re: [HACKERS] BEGIN inside transaction should be an error  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-patches
On 5/26/06, Alvaro Herrera <alvherre@commandprompt.com> wrote:
> Gurjeet Singh wrote:
> >    I wish to
> > know the standard procedure (command) to generate a patch; and from
> > which level in the source directory should I execute it?
>
> The toplevel directory.  The command is
>
> cvs -q diff -cp
>
> If you created new files to implement a patch, include them separately,
> indicating in the patch description in which directory they are meant to
> reside.

Thanks.

Here's the patch:

Index: src/backend/access/transam/xact.c
===================================================================
RCS file: /projects/cvsroot/pgsql/src/backend/access/transam/xact.c,v
retrieving revision 1.220
diff -c -p -r1.220 xact.c
*** src/backend/access/transam/xact.c    25 Apr 2006 00:25:17 -0000    1.220
--- src/backend/access/transam/xact.c    21 May 2006 15:40:00 -0000
*************** bool        XactReadOnly;
*** 59,64 ****
--- 59,65 ----
  int            CommitDelay = 0;    /* precommit delay in microseconds */
  int            CommitSiblings = 5; /* # concurrent xacts needed to sleep */

+ bool        BeginInXactIsError = true;

  /*
   *    transaction states - transaction state from server perspective
*************** BeginTransactionBlock(void)
*** 2725,2731 ****
          case TBLOCK_SUBINPROGRESS:
          case TBLOCK_ABORT:
          case TBLOCK_SUBABORT:
!             ereport(WARNING,
                      (errcode(ERRCODE_ACTIVE_SQL_TRANSACTION),
                       errmsg("there is already a transaction in progress")));
              break;
--- 2726,2732 ----
          case TBLOCK_SUBINPROGRESS:
          case TBLOCK_ABORT:
          case TBLOCK_SUBABORT:
!             ereport(ERROR,
                      (errcode(ERRCODE_ACTIVE_SQL_TRANSACTION),
                       errmsg("there is already a transaction in progress")));
              break;
Index: src/backend/utils/misc/guc.c
===================================================================
RCS file: /projects/cvsroot/pgsql/src/backend/utils/misc/guc.c,v
retrieving revision 1.318
diff -c -p -r1.318 guc.c
*** src/backend/utils/misc/guc.c    2 May 2006 11:28:55 -0000    1.318
--- src/backend/utils/misc/guc.c    21 May 2006 15:14:22 -0000
***************
*** 65,70 ****
--- 65,71 ----
  #include "utils/pg_locale.h"
  #include "pgstat.h"
  #include "access/gin.h"
+ #include "access/xact.h"

  #ifndef PG_KRB_SRVTAB
  #define PG_KRB_SRVTAB ""
*************** static struct config_bool ConfigureNames
*** 1005,1010 ****
--- 1006,1021 ----
          false, NULL, NULL
      },

+     {
+         {"begin_inside_transaction_is_error", PGC_USERSET, COMPAT_OPTIONS,
+          gettext_noop("A BEGIN statement inside a transaction raises ERROR."),
+          gettext_noop("It is provided to catch buggy applications. "
+                       "Disable it to let the buggy application run.")
+         },
+         &BeginInXactIsError,
+         true, NULL, NULL
+     },
+
      /* End-of-list marker */
      {
          {NULL, 0, 0, NULL, NULL}, NULL, false, NULL, NULL
Index: src/include/access/xact.h
===================================================================
RCS file: /projects/cvsroot/pgsql/src/include/access/xact.h,v
retrieving revision 1.82
diff -c -p -r1.82 xact.h
*** src/include/access/xact.h    25 Apr 2006 00:25:19 -0000    1.82
--- src/include/access/xact.h    21 May 2006 15:13:10 -0000
*************** extern int    XactIsoLevel;
*** 41,46 ****
--- 41,49 ----
  extern bool DefaultXactReadOnly;
  extern bool XactReadOnly;

+ /* A BEGIN statement inside a transaction raises ERROR */
+ extern bool BeginInXactIsError;
+
  /*
   *    start- and end-of-transaction callbacks for dynamically loaded modules
   */

pgsql-patches by date:

Previous
From: Andrew Dunstan
Date:
Subject: Re: [HACKERS] BEGIN inside transaction should be an error
Next
From: "Gurjeet Singh"
Date:
Subject: Re: [HACKERS] BEGIN inside transaction should be an error