Peter Geoghegan <pg@heroku.com> writes:
> On Mon, Jan 20, 2014 at 12:55 PM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> BTW, I'm also thinking that the "detected_version" kluge is about ready
>> to collapse of its own weight, or at least is clearly going to break in
>> future. What we need to do is embed the API version in the C name of the
>> pg_stat_statements function instead.
> I agree that it isn't scalable.
Yeah. It was more or less tolerable as long as we were only using it in
connection with identifying the set of output columns, but using it to
identify the expected input arguments too seems darn rickety. I'll
refactor so there are separate C call points for each supported API
version. (Well, I guess it's too late to separate 1.0 and 1.1, but
we can make 1.2 and future versions separate.)
regards, tom lane