I've just been running some timings of my JSON-producing functions, in 
particular array_to_json, and comparing them with the current 
XML-producing functions. Here's a typical result:

    andrew=# explain analyse select array_to_json(array_agg(q),true)
    from (select * from pg_attribute) q;
                                                          QUERY PLAN
    ---------------------------------------------------------------------------------------------------------------------
      Aggregate  (cost=70.77..70.78 rows=1 width=203) (actual
    time=38.919..38.920 rows=1 loops=1)
        ->  Seq Scan on pg_attribute  (cost=0.00..65.01 rows=2301
    width=203) (actual time=0.007..1.454 rows=2253 loops=1)
      Total runtime: 39.300 ms
    (3 rows)

    Time: 62.753 ms
    andrew=# explain analyse select table_to_xml('pg_attribute',
    true,false,'');
                                            QUERY PLAN
    ----------------------------------------------------------------------------------------
      Result  (cost=0.00..0.26 rows=1 width=0) (actual
    time=519.170..526.737 rows=1 loops=1)
      Total runtime: 526.780 ms
    (2 rows)


As you can see, producing the JSON is a heck of a lot faster than 
producing the equivalent XML. I had thought it might be necessary for 
good performance to cache the type output info in the FunctionCallInfo 
structure, rather than fetch it for each Datum we output, but that 
doesn't seem to be so. For now I'm inclined not to proceed with that, 
and leave it as an optimization to be considered later if necessary. 
Thoughts?

cheers

andrew

Responses

• Re: JSON output functions. at 2012-02-02 09:35:55 from Abhijit Menon-Sen

pgsql-hackers by date

Next:	From: Jim Nasby	Date: 2012-02-02 00:33:47
	Subject: Re: foreign key locks, 2nd attempt
Previous:	From: Jim Nasby	Date: 2012-02-01 23:47:05
	Subject: Re: feature request - datum_compute_size and datum write_should be public