Re: JSONPATH documentation

Поиск
Список
Период
Сортировка
От Steven Pousty
Тема Re: JSONPATH documentation
Дата
Msg-id CAKmB1PHBYfXSe16ksasyT92TzG+VwFRF--Tuim17n=c_5CtPKw@mail.gmail.com
обсуждение исходный текст
Ответ на Re: JSONPATH documentation  (Alexander Korotkov <a.korotkov@postgrespro.ru>)
Список pgsql-hackers
Privet :D

On Mon, Sep 23, 2019 at 12:29 PM Alexander Korotkov <a.korotkov@postgrespro.ru> wrote:
Hi!

On Mon, Sep 23, 2019 at 10:10 PM Steven Pousty <steve.pousty@gmail.com> wrote:
> https://standards.iso.org/ittf/PubliclyAvailableStandards/c067367_ISO_IEC_TR_19075-6_2017.zip

Yes, this link looks good to me.  It's technical report, not standard
itself.  So, it may have some little divergences.  But it seems to be
the best free resource available, assuming standard itself isn't free.

Works for me if we can't find something better
 
> I am uncertain why JSONPath is considered part of the datatype any more so than string functions are considered part of the character datatype
> https://www.postgresql.org/docs/11/functions-string.html

Let me clarify my thoughts.  SQL-level functions jsonb_path_*() (table
9.49) are clearly not part of jsonpath datatype.  But jsonpath
accessors (table 8.25), functions (table 9.44) and operators (table
9.45) are used inside jsonpath value.  So, technically they are parts
of jsonpath datatype.


Yes but the only time I would use those  8.25, 9.44, and 9.45 is to just create a jsonpath whose main purpose is to query or filter JSONB.
As a continued analogy, I think we rightly do not discuss anything but creating and considerations when using character fields:
https://www.postgresql.org/docs/11/datatype-character.html

And then we have a separate page that talk about all the ways you can manipulate and filter character fields.

My feeling is that JSONPath is only included as a way to work with JSONB, not as requirement of JSONB. Therefore JSONPath documentation belongs with all the other ways we work with JSONB, not as part of the datatype definition.
JSONPath is important and complicated enough that it may warrant its own page, just not in the same page where we define JSON(B)

 
P.S.  We don't use top posting in mailing lists.  Please, use bottom
posting.  See https://en.wikipedia.org/wiki/Posting_style#Top-posting
for details.


Thanks for the very KIND etiquette correction -  I really appreciate you not flaming me.
 
Thanks
Steve

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Alexander Korotkov
Дата:
Сообщение: Re: JSONPATH documentation
Следующее
От: Alexander Korotkov
Дата:
Сообщение: Re: Psql patch to show access methods info