Thread: [PATCH] 4 coverity patches for ECPG

Here's some patches to clear up some of the most obvious problems with
ECPG. This is just scratching the surface, but it's good to start
somewhere:

- coverity_163.diff

Strange assumption about var. The rest of the code in the function
assumes var will not be NULL, yet this one line does not.

- coverity_170.diff

Assumption that PQresultErrorField will never return NULL.

- coverity_244.diff

The code allows con to be NULL, yet if it is, this line will segfault.

- coverity_252_253.diff

Move the PGTYPESnumeric_free outside the loop to avoid the double free
possibility.

- ecpgtest.patch

A patch which add a "make check" to the ecpg/test directory. I don't
expect this to be committed to CVS, but it's provided as an example of
the tests the above patches pass.
--
Martijn van Oosterhout   <kleptog@svana.org>   http://svana.org/kleptog/
> From each according to his ability. To each according to his ability to litigate.

On Sun, Apr 23, 2006 at 11:55:03PM +0200, Martijn van Oosterhout wrote:
> Here's some patches to clear up some of the most obvious problems with
> ECPG. This is just scratching the surface, but it's good to start
> somewhere:
> ...

I just committed all patches to ecpglib. Sorry for the delay but it
appears I haven't been subscribed to -patches. Now I am.

The changes to the test directory weren't committed because the SoC work
includes bringing those tests into the regression system, so I don't
want to change the code base during that work.

Michael
--
Michael Meskes
Email: Michael at Fam-Meskes dot De, Michael at Meskes dot (De|Com|Net|Org)
ICQ: 179140304, AIM/Yahoo: michaelmeskes, Jabber: meskes@jabber.org
Go SF 49ers! Go Rhein Fire! Use Debian GNU/Linux! Use PostgreSQL!