Re: PL/Python: Fix return in the middle of PG_TRY() block.

On Mon, Jan 16, 2023 at 11:29 PM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:

> Xing Guo <higuoxing(at)gmail(dot)com> writes:
> > Are there any unsafe codes in pltcl.c? The return statement is in the
> > PG_CATCH() block, I think the exception stack has been recovered in
> > PG_CATCH block so the return statement in PG_CATCH block should be ok?
>
> Yes, the stack has already been unwound at the start of a PG_CATCH
> (or PG_FINALLY) block, so there is no reason to avoid returning
> out of those.
>
> In principle you could also mess things up with a "continue", "break",
> or "goto" leading out of PG_TRY. That's probably far less likely
> than "return", but I wonder whether Andres' compiler hack will
> catch that.
>
> regards, tom lane
>

Thank you Tom,

Based on your comments, I've refactored my clang checker[1], now it can
warn about the following patterns:
1. return statement in PG_TRY(). We've catched all of them in this thread.
2. continue statement in PG_TRY() *unless* it's in for/while/do-while
statements.
3. break statement in PG_TRY() *unless* it's in for/while/do-while/switch
statements.
4. goto statement in PG_TRY() *unless* the label it points to is in the
same PG_TRY block.

Good news is that, there's no patterns (2, 3, 4) in Postgres source tree
and we've catched all of the return statements in the PG_TRY block in this
thread.

[1]
https://github.com/higuoxing/clang-plugins/blob/main/lib/ReturnInPgTryBlockChecker.cpp

Best Regards,
Xing