Re: Can we go beyond the standard to make Postgres radically better?

Поиск
Список
Период
Сортировка
От Merlin Moncure
Тема Re: Can we go beyond the standard to make Postgres radically better?
Дата
Msg-id CAHyXU0wrJESzYewdeEMz9SFGVLHtJUYkhCyxoUu2++tSb_K8jg@mail.gmail.com
обсуждение исходный текст
Ответ на Re: Can we go beyond the standard to make Postgres radically better?  (Pavel Stehule <pavel.stehule@gmail.com>)
Ответы Re: Can we go beyond the standard to make Postgres radically better?  (Bruce Momjian <bruce@momjian.us>)
Список pgsql-general
On Sun, Feb 13, 2022 at 4:00 AM Pavel Stehule <pavel.stehule@gmail.com> wrote:
>
>
>
> ne 13. 2. 2022 v 10:45 odesílatel Guyren Howe <guyren@gmail.com> napsal:
>>
>>
>> The MySQL autocomplete is designed without context filtering. Maybe we can have this implementation too (as
alternative)
>>
>> so using all column names + all table names + aliases.column names (when we know defined alias)
>>
>> Another idea about column excluding. Any implementation on the client side is very complex, because you need to
parsesql. But maybe we can enhance SQL with some syntax. 
>>
>> SELECT ... and LAST optional clause can be HIDE, OR EXCLUDE NAME or EXCLUDE TYPE
>>
>> SELECT * FROM postgistab HIDE TYPE LIKE 'wkb%'
>> SELECT * FROM postgistab ORDER BY ... EXCLUDE COLUMN LIKE 'shape%'
>>
>> WITH x AS (SELECT * FROM xx)
>> SELECT * FROM x EXCLUDE COLUMN x1,x2
>>
>> The column excluding should be separate *last* clase.
>>
>> More with this syntax is less possibility of collision with ANSI SQL
>>
>> Not against this. Seems somewhere in here might be a nice quality of life change.
>>
>> Still.
>>
>> I originally suggested that SQL is terrible and we should fearlessly either replace it or migrate it toward
somethingbetter. And the thread winds up with a debate about a minor change to a development tool. 
>>
>> I’m back to just having no earthly idea why anyone who finds relations to be a productive tool for building a model
wouldthink that SQL being the only means to do that is Okay. 
>
> I think the rating of data langues is very subjective, and I am happy with SQL more than with special languages like
Dor Quel, or other. I know SQL has a lot of disadvantages, but it was designed for humans and it works for me. 

Exactly. SQL is proven to be more productive and code written in it
has longer longevity than alternatives.  It's also generally more
terse in the hands of a good author.  The authors of all the 'SQL
sucks' rants don't really explore why this is the case.  For example,
SQL has transactions and pretty much all other major languages don't.
They may have it in a limited sense but not standardized throughout
the syntax and the standard libraries.  High quality automatic
concurrency models are another factor.


merlin



В списке pgsql-general по дате отправления:

Предыдущее
От: Glen Eustace
Дата:
Сообщение: Re: Moving the master to a new server
Следующее
От: Bruce Momjian
Дата:
Сообщение: Re: Can we go beyond the standard to make Postgres radically better?