Re: jsonpath

Поиск
Список
Период
Сортировка
От Andrew Dunstan
Тема Re: jsonpath
Дата
Msg-id f0b44fe7-d47a-eff2-592c-d9dbe4a63f75@2ndQuadrant.com
обсуждение исходный текст
Ответ на Re: jsonpath  (Andrew Dunstan <andrew.dunstan@2ndquadrant.com>)
Ответы Re: jsonpath
Список pgsql-hackers

On 01/15/2018 07:24 PM, Andrew Dunstan wrote:
>
> On 01/10/2018 05:42 PM, Nikita Glukhov wrote:
>> Attached new 8th version of jsonpath related patches. Complete
>> documentation is still missing.
>>
>> The first 4 small patches are necessary datetime handling in jsonpath:
>> 1. simple refactoring, extracted function that will be used later in
>> jsonpath
>> 2. throw an error when the input or format string contains trailing
>> elements
>> 3. avoid unnecessary cstring to text conversions
>> 4. add function for automatic datetime type recognition by the
>> presence of formatting components
>>
>> Should they be posted in a separate thread?
>>
>
> The first of these refactors the json/jsonb timestamp formatting into a
> single function, removing a lot of code duplication. The involves
> exposing time2tm() and timetz2tm(). I don't think that's a tragedy, so
> unless there is any objection I propose to commit it shortly.
>
> The next three expose a bit more of the date/time API. I'm still
> reviewing those.
>


I have committed the first of these patches.

I have reviewed the next three, and I think they are generally good.
There is no real point in committing them ahead of the jsonpath patch
since there would be no point in having them at all but for that patch.

Note that these do export the following hitherto internal bits of the
datetime functionality:

    tm2time
    tm2timetz
    AdjustTimeForTypmod
    AdjustTimestampForTypmod

Moving on to review the main jsonpath patch.

cheers

andrew

-- 
Andrew Dunstan                https://www.2ndQuadrant.com
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services



В списке pgsql-hackers по дате отправления:

Предыдущее
От: Tom Lane
Дата:
Сообщение: Re: PostgreSQL crashes with SIGSEGV
Следующее
От: Robert Haas
Дата:
Сообщение: Re: [HACKERS] postgres_fdw bug in 9.6