| From: | Robert Haas <robertmhaas(at)gmail(dot)com> | 
|---|---|
| To: | "pgsql-hackers(at)postgresql(dot)org" <pgsql-hackers(at)postgresql(dot)org> | 
| Subject: | ALTER TABLE .. ADD PRIMARY KEY .. USING INDEX has dump-restore hazard | 
| Date: | 2015-07-21 16:23:47 | 
| Message-ID: | CA+TgmoYMsEPePx6J2oQHRqRmZuSQevFBbabyji6To1NK3TUs1Q@mail.gmail.com | 
| Views: | Whole Thread | Raw Message | Download mbox | Resend email | 
| Thread: | |
| Lists: | pgsql-hackers | 
Consider:
rhaas=# create table foo (a text);
CREATE TABLE
rhaas=# create unique index on foo (a collate "C");
CREATE INDEX
rhaas=# alter table foo add primary key using index foo_a_idx;
ALTER TABLE
rhaas=# \d foo
    Table "public.foo"
 Column | Type | Modifiers
--------+------+-----------
 a      | text | not null
Indexes:
    "foo_a_idx" PRIMARY KEY, btree (a COLLATE "C")
Now dump and restore this database. Then:
rhaas=# \d foo
    Table "public.foo"
 Column | Type | Modifiers
--------+------+-----------
 a      | text | not null
Indexes:
    "foo_a_idx" PRIMARY KEY, btree (a)
Notice that the collation specifier is gone. Oops.
-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Alexander Korotkov | 2015-07-21 16:28:01 | Re: Selectivity estimation for intarray with @@ | 
| Previous Message | Heikki Linnakangas | 2015-07-21 16:20:12 | Re: Fillfactor for GIN indexes |