Amit Langote <amitlangote09@gmail.com> writes:
> On Fri, Jan 26, 2024 at 0:15 Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> Amit Langote <amitlangote09@gmail.com> writes:
>>> Ignoring the warning was my 1st thought too, because an old discussion I
>>> found about the warning was too old (2011). The style I adopted in my
>>> “fix” is used in a few other places too, so I thought I might as well
>>> go for it.
>> Oh, well maybe I'm missing some context. What comparable places did
>> you see?
> Sorry, not on my computer right now so can’t paste any code, but I was able
> to find a couple of functions (using Google ;) that refer to error_occurred
> directly for one reason or another:
OK, looking around, it seems like our pattern is that direct access
to escontext.error_occurred is okay in the same function that sets up
the escontext (so that there's no possibility of a NULL pointer).
So this change is fine and I withdraw my objection. Sorry for
the noise.
regards, tom lane