lo_creat() bug - Mailing list pgsql-hackers

From Gavin Sherry
Subject lo_creat() bug
Date
Msg-id Pine.LNX.4.21.0109171818430.12374-100000@linuxworld.com.au
Whole thread Raw
Responses Re: lo_creat() bug  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
Hi all,

I was just looking through libpq large object code an noticed what seemed
to be a bug. The function lo_create() is declared as Oid yet it contains
the following code:
   if (conn->lobjfuncs == (PGlobjfuncs *) NULL)   {       if (lo_initialize(conn) < 0)           return -1;   }

If lo_initialize returns < 0, you have some pretty serious problems - out
of memory, conn is invalid, etc. However, casting -1 to Oid returns what
seems to be a valid Oid. Shouldn't it return InvalidOid?

Gavin





pgsql-hackers by date:

Previous
From: Thomas Lockhart
Date:
Subject: Re: [Fwd: [Fwd: [tao-users] FW: HEADS UP: CVSup timestamp bug]]
Next
From: Turbo Fredriksson
Date:
Subject: Hot spare PSQL server