On Sun, Jan 20, 2013 at 10:01 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
> Robert Haas <robertmhaas@gmail.com> writes:
>> On Sun, Jan 20, 2013 at 2:26 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>>> Pavel is claiming it's okay for that to fall over if the array has
>>> more than 100 elements. I disagree, not only for the specific case of
>>> CONCAT(), but with the more general implication that such a limitation
>>> is going to be okay for any VARIADIC ANY function that anyone will ever
>>> write.
>
>> I don't know - how many of those will there really ever be? I mean,
>> people only write functions as VARIADIC as a notational convenience,
>> don't they? If you actually need to pass more than 100 separate
>> pieces of data to a function, sending over 100+ parameters is almost
>> certainly the Wrong Way To Do It.
>
> Well, not necessarily, if they're reasonably expressed as an array.
> I would also point out that there is no corresponding limitation on
> variadic functions that take any type other than ANY. Indeed, despite
> Pavel's claim to the contrary, I'm pretty sure it's seen as a feature
> that there's no specific upper limit to how many parameters you can pass
> to a variadic function when using the "VARIADIC array-value" syntax.
> It's certainly a feature that you can pass a varying number of
> parameters that way, thereby "evading" the syntactic fact that you can't
> pass a varying number of parameters any other way. I don't see how
> come it isn't a feature that you can "evade" the FUNC_MAX_ARGS limit
> that way, or why we'd consider it acceptable for variably-sized
> parameter arrays to have such a small arbitrary limit.
OK, I see. If people are already counting on there being no fixed
limit for variadic functions with a type other than "any", then it
would indeed seem weird to make "any" an exception. I'm not sure how
much practical use case there is for such a thing, but still.
--
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company