Re: Question about GetAttributeByNum(Name) ExecQual.c
От | Zdenek Kotala |
---|---|
Тема | Re: Question about GetAttributeByNum(Name) ExecQual.c |
Дата | |
Msg-id | 4909ADAD.7080902@sun.com обсуждение исходный текст |
Ответ на | Re: Question about GetAttributeByNum(Name) ExecQual.c (Tom Lane <tgl@sss.pgh.pa.us>) |
Ответы |
Re: Question about GetAttributeByNum(Name) ExecQual.c
|
Список | pgsql-hackers |
Tom Lane napsal(a): > Zdenek Kotala <Zdenek.Kotala@Sun.COM> writes: >> GetAttributeByNum has first parameter as HeapTupleHeader, but most functions >> use Datum and after that they call DatumGetHeapTupleHeader. The difference is >> that DatumGetHeapTupleHeader performs detoast on tuple(row type). > >> Is it intention do not detoast in these functions or it is a bug? > > You would certainly not want a tuple to get separately detoasted for > each attribute you pull from it. So having detoasting here would be > the wrong thing IMHO. Does it mean that these function are every time called with heap tuple or untoasted row type (composite data type)? What is purpose of these function. They are not use in executor and never in the core (only in example and regress). Should be really in executor? Thanks Zdenek -- Zdenek Kotala Sun Microsystems Prague, Czech Republic http://sun.com/postgresql
В списке pgsql-hackers по дате отправления: