Re: Minor improvements in alter_table.sgml

(2014/04/09 12:03), Etsuro Fujita wrote:
> (2014/04/09 1:23), Robert Haas wrote:
>> On Tue, Apr 8, 2014 at 5:05 AM, Etsuro Fujita
>> <fujita(dot)etsuro(at)lab(dot)ntt(dot)co(dot)jp> wrote:
>>> Attached is a patch to improve the manual page for the ALTER TABLE
>>> command.
>>
>> Do we really need to add a section for "type_name" when we already
>> have a section for "OF type_name"?
>
> I think that the section for "type_name" would be necessary as that in
> chapter "Parameters", not in chapter "Description", which includes the
> section for "OF type_name".
>
>> constraint_name is also used for adding a constraint using an index.
>> So it could not only be a constraint to alter, validate, or drop, but
>> also a new constraint name to be added.
>
> I overlooked that.
>
> > Honestly, how much value is
>> there in even having a section for this? Do we really want to
>> document constraint_name as "name of an existing constraint, or the
>> name of a new constraint to be added"? It would be accurate, then,
>> but it also doesn't really tell you anything you didn't know already.
>
> You have a point there, but I feel odd about the documentation as is,
> because some are well written (eg, column_name) and some are not (eg,
> constraint_name). So, if there are no objections, I'd like to update
> the patch.

Attached is an updated version of the patch.

Thanks,

Best regards,
Etsuro Fujita