Date: Thu, 12 Jan 2023 10:44:33 -0800
From: Nathan Bossart <nathandbossart@gmail.com>
To: Xing Guo <higuoxing@gmail.com>
Cc: pgsql-hackers@postgresql.org
Subject: Re: PL/Python: Fix return in the middle of PG_TRY() block.
Message-ID: <20230112184433.GA2104952@nathanxps13>
References: 
 <CACpMh+CMsGMRKFzFMm3bYTzQmMU5nfEEoEDU2apJcc4hid36AQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: 
 <CACpMh+CMsGMRKFzFMm3bYTzQmMU5nfEEoEDU2apJcc4hid36AQ@mail.gmail.com>
Archived-At: 
 <https://www.postgresql.org/message-id/20230112184433.GA2104952%40nathanxps13>
Precedence: bulk

On Thu, Jan 12, 2023 at 11:19:29PM +0800, Xing Guo wrote:
> I was running static analyser against PostgreSQL and found there're 2
> return statements in PL/Python module which is not safe. Patch is
> attached.

Is the problem that PG_exception_stack and error_context_stack aren't
properly reset?

> @@ -690,12 +690,12 @@ PLy_trigger_build_args(FunctionCallInfo fcinfo, PLyProcedure *proc, HeapTuple *r
>  	PyObject   *volatile pltdata = NULL;
>  	char	   *stroid;
>  
> +	pltdata = PyDict_New();
> +	if (!pltdata)
> +		return NULL;
> +
>  	PG_TRY();
>  	{
> -		pltdata = PyDict_New();
> -		if (!pltdata)
> -			return NULL;
> -
>  		pltname = PLyUnicode_FromString(tdata->tg_trigger->tgname);
>  		PyDict_SetItemString(pltdata, "name", pltname);
>  		Py_DECREF(pltname);

There's another "return" later on in this PG_TRY block.  I wonder if it's
possible to detect this sort of thing at compile time.

-- 
Nathan Bossart
Amazon Web Services: https://aws.amazon.com