| From: | Craig Ringer <craig(at)2ndquadrant(dot)com> |
|---|---|
| To: | Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com> |
| Cc: | PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org> |
| Subject: | Re: patch: function xmltable |
| Date: | 2016-09-07 07:23:59 |
| Message-ID: | CAMsr+YE0xPpPBw8vfL0LabtL_Ry30KOPWWtR84h5DFV1qOXoHQ@mail.gmail.com |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On 7 September 2016 at 14:44, Pavel Stehule <pavel(dot)stehule(at)gmail(dot)com> wrote:
>>
>> Suggested comment:
>>
>> /*
>> * This is the parsenode for a column definition in a table-expression
>> like XMLTABLE.
>> *
>> * We can't re-use ColumnDef here; the utility command column
>> definition has all the
>> * wrong attributes for use in table-expressions and just doesn't make
>> sense here.
>> */
>> typedef struct TableExprColumn
>> {
>> ...
>> };
>>
>> ?
>>
>> Why "RawCol" ? What does it become when it's not "raw" anymore? Is
>> that a reference to ColumnDef's raw_default and cooked_default for
>> untransformed vs transformed parse-trees?
>
>
> My previous reply was wrong - it is used by parser only and holds TypeName
> field. The analogy with ColumnDef raw_default is perfect.
Cool, lets just comment that then.
I'll wait on an updated patch per discussion to date. Hopefully
somebody else with more of a clue than me can offer better review of
the executor/view/caching part you specifically called out as complex.
Otherwise maybe it'll be clearer in a revised version.
--
Craig Ringer http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Training & Services
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tsunakawa, Takayuki | 2016-09-07 07:39:09 | Re: Supporting SJIS as a database encoding |
| Previous Message | Kyotaro HORIGUCHI | 2016-09-07 07:13:04 | Re: Supporting SJIS as a database encoding |