Silly Newbie question

Aaron Spiteri wrote:

To all,

I am brand new to writing functions for PostgreSQL and understand that
this question has been asked one hundred and one times. Please be
patient with me, I do learn quickly.

My question is:

I wish to create a an aggregate function and have read and re-read the
34.10 section of the 8.3 documentation as well as scanned through the
fmgr header file. My problem is that I know certain structures are
created with the idea of retaining information for use in the state
function but I don't really understand which structures I should be
looking at and how to initilize them (if needed) in the sfunc. Any
pushes in the right direction would be very appreciated, I am using
the Version 1 calling convention.

<http://a.ninemsn.com.au/b.aspx?URL=http%3A%2F%2Fsecure%2Dau%2Eimrworldwide%2Ecom%2Fcgi%2Dbin%2Fa%2Fci%5F450304%2Fet%5F2%2Fcg%5F801459%2Fpi%5F1004813%2Fai%5F859641&amp;_t=762955845&amp;_r=tig_OCT07&amp;_m=EXT&gt;

This is not the correct forum for such questions. Please ask again on
pgsql-general. pgsql-hackers is about development of postgresql itself.

cheers

andrew

Aaron Spiteri <azzmosphere@hotmail.com> writes:

I wish to create a an aggregate function and have read and re-read the
34.10 section of the 8.3 documentation as well as scanned through the
fmgr header file. My problem is that I know certain structures are
created with the idea of retaining information for use in the state
function but I don't really understand which structures I should be
looking at and how to initilize them (if needed) in the sfunc. Any
pushes in the right direction would be very appreciated, I am using
the Version 1 calling convention.

Are there any existing aggregates that do something similar to what you
need? If so, try taking a look at their source code. The general rule
in writing C code for Postgres is that it's an open-book test ;-) ---
you are expected to look at the code, not only the documentation.

regards, tom lane

Andrew Dunstan <andrew@dunslane.net> writes:

Aaron Spiteri wrote:

I wish to create a an aggregate function and have read and re-read the
34.10 section of the 8.3 documentation as well as scanned through the
fmgr header file.

This is not the correct forum for such questions. Please ask again on
pgsql-general. pgsql-hackers is about development of postgresql itself.

In the past we've encouraged people who are writing C-code additions
to discuss on -hackers. I'm not sure that -general is a better place
for that; what fraction of -general readers do you think are interested
or qualified to help?

I recall seeing similar questions lately on -novice, where they
*surely* seem out of place. If we've scared people away from -hackers
to the extent that they post C-code-hacking questions on -novice,
I think there's something wrong...

regards, tom lane

Tom Lane wrote:

Andrew Dunstan <andrew@dunslane.net> writes:

Aaron Spiteri wrote:

I wish to create a an aggregate function and have read and re-read the
34.10 section of the 8.3 documentation as well as scanned through the
fmgr header file.

This is not the correct forum for such questions. Please ask again on
pgsql-general. pgsql-hackers is about development of postgresql itself.

In the past we've encouraged people who are writing C-code additions
to discuss on -hackers. I'm not sure that -general is a better place
for that; what fraction of -general readers do you think are interested
or qualified to help?

I recall seeing similar questions lately on -novice, where they
*surely* seem out of place. If we've scared people away from -hackers
to the extent that they post C-code-hacking questions on -novice,
I think there's something wrong...

OK, fair enough.

cheers

andrew