Re: [PATCH] Store Extension Options

Поиск
Список
Период
Сортировка
От Fabrízio de Royes Mello
Тема Re: [PATCH] Store Extension Options
Дата
Msg-id CAFcNs+p+2OA2fg7o-8KWmckazjAYWue6mVNnUdpuRpT0PZ8D_g@mail.gmail.com
обсуждение исходный текст
Ответ на Re: [PATCH] Store Extension Options  (Robert Haas <robertmhaas@gmail.com>)
Список pgsql-hackers
<div dir="ltr"><div class="gmail_extra"><br />On Thu, Mar 13, 2014 at 5:35 PM, Robert Haas <<a
href="mailto:robertmhaas@gmail.com">robertmhaas@gmail.com</a>>wrote:<br />><br />> Well, I'm not sure that's
reallyany big deal, but I'm not wedded to<br /> > the label idea.  My principal concern is: I'm opposed to
allowing<br/>> unvalidated options into the database.  I think it should be a<br />> requirement that if the
validatorcan't be found and called, then the<br /> > reloption is no good and you just reject it.  So, if we go with
the<br/>> reloptions route, I'd want to see pg_register_option_namespace go away<br />> in favor of some solution
thatpreserves that property.  One thing I<br /> > kind of like about the LABEL approach is that it applies to
virtually<br/>> every object type, meaning that we might not have to repeat this<br />> discussion when (as seems
inevitable)people want to be able to do<br /> > things to schemas or tablespaces or roles.  But I don't hold that<br
/>>position so strongly as to be unwilling to entertain any other<br />> options.<br />><br /><br />During the
lastdays I thought about this discussion and to use SECLABELs sounds weird to me. Here in Brazil we call this kind of
thing'gambiarra'. Because we'll try to use something that born with a very well defined purpose to another purpose.
PersonallyI don't like that.<br /><br />If we think more about SECLABELs, in a more abstract way, it is just a
'property'about database objects. And the same is COMMENTs. Both SECLABEL and COMMENT provide a way to store something
aboutobjects.<br /><br />Maybe we can think about a new object on top of COMMENT and SECLABELs. An object called
'PROPERTY'or 'OPTION'. And COMMENTs and SECLABELs can be just a kind of this new object, and we must introduce a new
kindcallled 'CUSTOM'.<br /><br />I thought about this because representation (syntax) of SECLABELs and COMMENTs are
similar.They describe something about objects, they have local and shared catalog.<br /><br />This is just something
thatoccurred to me. Maybe I'm completely wrong. Or not!<br /><br /></div><div class="gmail_extra">Comments?<br /><br
/></div><divclass="gmail_extra">--<br />Fabrízio de Royes Mello<br />Consultoria/Coaching PostgreSQL<br />>>
Timbira:<a href="http://www.timbira.com.br">http://www.timbira.com.br</a><br /> >> Blog sobre TI: <a
href="http://fabriziomello.blogspot.com">http://fabriziomello.blogspot.com</a><br/>>> Perfil Linkedin: <a
href="http://br.linkedin.com/in/fabriziomello">http://br.linkedin.com/in/fabriziomello</a><br/> >> Twitter: <a
href="http://twitter.com/fabriziomello">http://twitter.com/fabriziomello</a></div></div>

В списке pgsql-hackers по дате отправления:

Предыдущее
От: Heikki Linnakangas
Дата:
Сообщение: Re: Only first XLogRecData is visible to rm_desc with WAL_DEBUG
Следующее
От: Robert Haas
Дата:
Сообщение: Re: Only first XLogRecData is visible to rm_desc with WAL_DEBUG