Обсуждение: removal of types

I would like to remove the following types in 6.4:

    oidint2
    oidint4
    oidname

These store composite values, i.e., stores an oid and an int2.  Not sure
why these ever existed, and I would be surprised if anyone used them.

Comments?

--
Bruce Momjian                          |  830 Blythe Avenue
maillist@candle.pha.pa.us              |  Drexel Hill, Pennsylvania 19026
  +  If your life is a hard drive,     |  (610) 353-9879(w)
  +  Christ can be your backup.        |  (610) 853-3000(h)

Bruce Momjian wrote:
>
> I would like to remove the following types in 6.4:
>
>         oidint2
>         oidint4
>         oidname
>
> These store composite values, i.e., stores an oid and an int2.  Not sure
> why these ever existed, and I would be surprised if anyone used them.

local1_template1.bki.source:

declare index pg_attribute_relid_attnam_index on pg_attribute
        using btree(mkoidname(attrelid, attname) oidname_ops)
                                                 ^^^^^^^
declare index pg_attribute_relid_attnum_index on pg_attribute
        using btree(mkoidint2(attrelid, attnum) oidint2_ops)
                                                ^^^^^^^
Actually, having multi-key indices we could use them
istead of functional indices...

Vadim

> Bruce Momjian wrote:
> >
> > I would like to remove the following types in 6.4:
> >
> >         oidint2
> >         oidint4
> >         oidname
> >
> > These store composite values, i.e., stores an oid and an int2.  Not sure
> > why these ever existed, and I would be surprised if anyone used them.
>
> local1_template1.bki.source:
>
> declare index pg_attribute_relid_attnam_index on pg_attribute
>         using btree(mkoidname(attrelid, attname) oidname_ops)
>                                                  ^^^^^^^
> declare index pg_attribute_relid_attnum_index on pg_attribute
>         using btree(mkoidint2(attrelid, attnum) oidint2_ops)
>                                                 ^^^^^^^
> Actually, having multi-key indices we could use them
> istead of functional indices...

Yes.  I would have to add multi-key functionality the bootstrap process.


--
Bruce Momjian                          |  830 Blythe Avenue
maillist@candle.pha.pa.us              |  Drexel Hill, Pennsylvania 19026
  +  If your life is a hard drive,     |  (610) 353-9879(w)
  +  Christ can be your backup.        |  (610) 853-3000(h)