Hi,
At Wed, 19 Feb 2014 16:17:05 +0900, Shigeru Hanada wrote
> 2014-02-18 19:29 GMT+09:00 Kyotaro HORIGUCHI <horiguchi.kyotaro@lab.ntt.co.jp>:
> > Could you guess any use cases in which we are happy with ALTER
> > TABLE's inheritance tree walking? IMHO, ALTER FOREIGN TABLE
> > always comes with some changes of the data source so implicitly
> > invoking of such commands should be defaultly turned off.
>
> Imagine a case that foreign data source have attributes (A, B, C, D)
> but foreign tables and their parent ware defined as (A, B, C). If
> user wants to use D as well, ALTER TABLE parent ADD COLUMN D type
> would be useful (rather necessary?) to keep consistency.
Hmm. I seems to me an issue of mis-configuration at first
step. However, my anxiety is - as in my message just before -
ALTER'ing foreign table definitions without any notice to
operatos and irregular or random logic on check applicability(?)
of ALTER actions.
> Changing data type from compatible one (i.e., int to numeric,
> varchar(n) to text), adding CHECK/NOT NULL constraint would be also
> possible.
I see, thank you. Changing data types are surely valuable but
also seems difficult to check validity:(
Anyway, I gave a second thought on this issue. Please have a look
on that.
regards,
--
Kyotaro Horiguchi
NTT Open Source Software Center