On Thu, Nov 03, 2005 at 03:12:33PM +0000, Bernhard Weisshuhn <[EMAIL
PROTECTED]> wrote:
> The following bug has been logged online:
>
> Bug reference: 2019
> Logged by: Bernhard Weisshuhn
> Email address: [EMAIL PROTECTED]
> PostgreSQL version: 8.1rc1
> Operating system: Linux Fedora Core 4 x86_64
> Description: tsearch2-related coredump
Teodor Sigaev sent me a patch even before the web submission got past
the moderators! Not only is tsearch2 quick, its developers are, too!
;-)
The attached patch for Teodor seems to fix the problem for me. I say
'seems' since my testings conditions differ a bit from the ones the
problem appeared, but I did test the same set of queries.
cheers everybody,
Bernhard Weißhuhn
*** postgresql-8.1RC1/contrib/tsearch2/ispell/regis.h.orig Thu Nov 3
20:41:27 2005
--- postgresql-8.1RC1/contrib/tsearch2/ispell/regis.h Thu Nov 3 21:06:48 2005
***************
*** 13,19 ****
unsigned char data[1];
} RegisNode;
! #define RNHDRSZ (sizeof(uint32)+sizeof(void*))
#define RSF_ONEOF 1
#define RSF_NONEOF 2
--- 13,19 ----
unsigned char data[1];
} RegisNode;
! #define RNHDRSZ (offsetof(RegisNode,data))
#define RSF_ONEOF 1
#define RSF_NONEOF 2
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
