Re: [HACKERS] Bug in pg_dump --table and --exclude-table fordeclarative partition table handling.

On Tue, May 9, 2017 at 3:13 PM, Amit Langote
<Langote_Amit_f8@lab.ntt.co.jp> wrote:
> On 2017/05/09 17:21, Jeevan Ladhe wrote:
>> On Tue, May 9, 2017 at 12:43 PM, amul sul <sulamul@gmail.com> wrote:
>>> Current pg_dump --exclude-table option excludes partitioned relation
>>> and dumps all its child relations and vice versa for --table option, which
>>> I think is incorrect.
>>>
>>> In this case we might need to explore all partitions and exclude or include
>>> from dump according to given pg_dump option, attaching WIP patch proposing
>>> same fix.   Thoughts/Comments?
>>
>> Also, I can see similar issue exists with inheritance.
>> In attached patch, I have extended Amul's original patch to address the
>> inheritance dumping issue.
>
> Perhaps, it will be better not to touch the regular inheritance tables in
> this patch.

Yeah, I think it's fine if parent gets dumped without one or more of
its children, that's user's choice when it used a certain pattern.
Problematic case might be when we dump a child without its parent and
have INHERITS clause there. pg_restore would throw an error. But in
case that problem exists it's very old and should be fixed separately.