Re: RfD: more powerful "any" types

On Fri, Sep 11, 2009 at 11:19 AM, Robert Haas <robertmhaas@gmail.com> wrote:
> On Fri, Sep 11, 2009 at 10:30 AM, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> "Kevin Grittner" <Kevin.Grittner@wicourts.gov> writes:
>>> I think the main benefit of a sprintf type function for PostgreSQL is
>>> in the formatting (setting length, scale, alignment), not in making
>>> concatenation more pretty.
>>
>> Exactly, which is why I'm so distressed that this proposal not only
>> hasn't got that, but is designed so that it's impossible to add it
>> later.
>
> I like the idea of making concatenation more pretty, quite frankly.
> No one has really responded to Pavel's contention that this is what
> to_char() is for.  Twice the code paths = twice the bugs, twice the
> places that have to be updated when some new feature is added, etc.

If you are going to use printf format codes, which is good and useful
being something of a standard, I'd call routine printf (not format)
and actually wrap vsnprintf.  The format codes in printf have a very
specific meaning: converting native C types to arrays of characters.
I think that a postgresql implementation should do exactly that:
attempt to convert the passed in datum to the c type in question if
possible (erroring if no cast exists) and then pass it down.  The idea
is we are not adding new formatting routines but using a very high
quality existing one...why reinvent the wheel?

so if you did: select printf('%s %3.1f', foo::box, bar::circle);
the box to char* cast would work (using the text cast) but the second
cast would fail unless the user added a cast to float.  The code in
question is easy to imagine...parse the format string, and loop the
varargs using the appropriate looked up cast one by one...

merlin