> "Christopher Kings-Lynne" <chriskl@familyhealth.com.au> writes:
> >> That was my first thought also, but then the wrong attnum would be used
> >> in the "make_var". Ugh. I think what Chris needs to do is extend the
> >> eref data structure so that there can be placeholders for dropped
> >> attributes. Perhaps NULLs could be included in the list, and then the
> >> code would become like
>
> > Hmmm... I don't get it - at the moment I'm preventing them from even
> > getting into the eref and all regression tests pass and every test I try
> > works as well...
>
> Are you checking access to columns that're to the right of the one
> dropped?
OK, interesting:
test=# create table test (a int4, b int4, c int4, d int4);
CREATE TABLE
test=# insert into test values (1,2,3,4);
INSERT 16588 1
test=# alter table test drop b;
ALTER TABLE
test=# select * from test;a | d | d
---+---+---1 | 3 | 4
(1 row)
It half works, half doesn't. Sigh - how come these things always turn out
harder than I think!?
pg_attribute:
test=# select attrelid,attname,attisdropped from pg_attribute where
attrelid=16586 and attnum > 0;attrelid | attname | attisdropped
----------+-----------+-------------- 16586 | a | f 16586 | dropped_2 | t 16586 | c | f 16586 |
d | f
(4 rows)
Chris