Обсуждение: Correct place for feature requests

Good day.

Is there any correct place for feature requests?

Thanks.

I think that PostgreSQL is great RDBMS, but one important (for me)
feature that it missing is case-insensitive identifier quotes
([table].[column]) like in Microsoft SQL Server.

I know that there are double quotes, but they are case-sensitive.

Or maybe some option that will make double quotes case-insensitive.

2015-06-25 20:03 GMT+03:00 Алексей Бережняк <core@irc.lg.ua>:
> Good day.
>
> Is there any correct place for feature requests?
>
> Thanks.

=?UTF-8?B?0JDQu9C10LrRgdC10Lkg0JHQtdGA0LXQttC90Y/Qug==?= <core@irc.lg.ua> writes:
> Is there any correct place for feature requests?

Starting a thread here is a reasonable way.

            regards, tom lane

On 6/25/2015 11:59 AM, Алексей Бережняк wrote:
> I think that PostgreSQL is great RDBMS, but one important (for me)
> feature that it missing is case-insensitive identifier quotes
> ([table].[column]) like in Microsoft SQL Server.

putting names in [ ]  is not part of any SQL specification.

> I know that there are double quotes, but they are case-sensitive.
>
> Or maybe some option that will make double quotes case-insensitive.

the current behavior is compliant with the SQL specification.   if you
want case-insensitive, don't quote the identifiers.   if you do quote
them, they are implicitly case sensitive according to the specifications.

--
john r pierce, recycling bits in santa cruz

John R Pierce wrote:
> On 6/25/2015 11:59 AM, Алексей Бережняк wrote:

> >Or maybe some option that will make double quotes case-insensitive.
>
> the current behavior is compliant with the SQL specification.   if you want
> case-insensitive, don't quote the identifiers.   if you do quote them, they
> are implicitly case sensitive according to the specifications.

What happens if you have spaces in your identifiers?

--
Álvaro Herrera                http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services

I understand that, but it seem there are no way to use words like
"group", or "case" without enclosing them into double quotes (and
therefore making them case-sensitive).

2015-06-25 22:29 GMT+03:00 John R Pierce <pierce@hogranch.com>:
> On 6/25/2015 11:59 AM, Алексей Бережняк wrote:
>>
>> I think that PostgreSQL is great RDBMS, but one important (for me)
>> feature that it missing is case-insensitive identifier quotes
>> ([table].[column]) like in Microsoft SQL Server.
>
>
> putting names in [ ]  is not part of any SQL specification.
>
>> I know that there are double quotes, but they are case-sensitive.
>>
>> Or maybe some option that will make double quotes case-insensitive.
>
>
> the current behavior is compliant with the SQL specification.   if you want
> case-insensitive, don't quote the identifiers.   if you do quote them, they
> are implicitly case sensitive according to the specifications.
>
> --
> john r pierce, recycling bits in santa cruz
>
>
>
>
> --
> Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
> To make changes to your subscription:
> http://www.postgresql.org/mailpref/pgsql-general

On 6/25/2015 12:33 PM, Alvaro Herrera wrote:
> What happens if you have spaces in your identifiers?


either don't do that, or use quoted identifiers and be consistent about
your case.

what if you have spaces in your identifiers in your
java/c/python/perl/php/basic/fortran/etc program?  oh wait, you can't DO
that.




--
john r pierce, recycling bits in santa cruz

On 06/25/2015 12:52 PM, Алексей Бережняк wrote:
> I understand that, but it seem there are no way to use words like
> "group", or "case" without enclosing them into double quotes (and
> therefore making them case-sensitive).

 From here:

https://technet.microsoft.com/en-us/library/ms176027%28v=sql.105%29.aspx
"
Delimited identifiers are used in the following situations:

     When reserved words are used for object names or parts of object names.

     Reserved keywords should not be used as object names. Databases
upgraded from earlier versions of SQL Server may contain identifiers
that include words not reserved in the earlier version, but that are
reserved words for the current version of SQL Server. You can refer to
the object by using delimited identifiers until the name can be changed. "

So in SQL Server using keywords is not considered a good idea.

From here:

https://msdn.microsoft.com/en-us/library/ms144250%28v=sql.105%29.aspx

and here:

https://msdn.microsoft.com/en-us/library/ms175874%28v=sql.100%29.aspx

it would seem case sensitivity is a determined by collation, as much as
quoting.

>
> 2015-06-25 22:29 GMT+03:00 John R Pierce <pierce@hogranch.com>:
>> On 6/25/2015 11:59 AM, Алексей Бережняк wrote:
>>>
>>> I think that PostgreSQL is great RDBMS, but one important (for me)
>>> feature that it missing is case-insensitive identifier quotes
>>> ([table].[column]) like in Microsoft SQL Server.
>>
>>
>> putting names in [ ]  is not part of any SQL specification.
>>
>>> I know that there are double quotes, but they are case-sensitive.
>>>
>>> Or maybe some option that will make double quotes case-insensitive.
>>
>>
>> the current behavior is compliant with the SQL specification.   if you want
>> case-insensitive, don't quote the identifiers.   if you do quote them, they
>> are implicitly case sensitive according to the specifications.
>>
>> --
>> john r pierce, recycling bits in santa cruz
>>
>>
>>
>>
>> --
>> Sent via pgsql-general mailing list (pgsql-general@postgresql.org)
>> To make changes to your subscription:
>> http://www.postgresql.org/mailpref/pgsql-general
>
>


--
Adrian Klaver
adrian.klaver@aklaver.com

On Thu, Jun 25, 2015 at 1:59 PM, Алексей Бережняк <core@irc.lg.ua> wrote:
> I think that PostgreSQL is great RDBMS, but one important (for me)
> feature that it missing is case-insensitive identifier quotes
> ([table].[column]) like in Microsoft SQL Server.
>
> I know that there are double quotes, but they are case-sensitive.
>
> Or maybe some option that will make double quotes case-insensitive.

This behavior is not going to change because of having to support
decades of code written around the current set of rules.  SQL Server
is very much in the same boat having inherited its somewhat quixotic
casing rules.  SQL server's semantics are not SQL standard anyways so
even if postgres did change it would not be in that direction.  SQL
server is basically fully case insensitive while postgres is case
sensitive (but works around this to some extent via case folding).

It's really somewhat of a shame, but to write portable SQL you want to
name_identifiers_like_this (particularly with postgres due to case
folding) and completely avoid syntax that requires identifier quoting.
  Also stay away from keywords or anything that might become one.

merlin