Re: pgAdmin III commit: Install adminpack on user request in frmStatus

Поиск
Список
Период
Сортировка
От Guillaume Lelarge
Тема Re: pgAdmin III commit: Install adminpack on user request in frmStatus
Дата
Msg-id 4DADF765.3030301@lelarge.info
обсуждение исходный текст
Ответ на Re: pgAdmin III commit: Install adminpack on user request in frmStatus  (Dave Page <dpage@pgadmin.org>)
Ответы Re: pgAdmin III commit: Install adminpack on user request in frmStatus  (Dave Page <dpage@pgadmin.org>)
Список pgadmin-hackers
Le 19/04/2011 12:47, Dave Page a écrit :
> On Tue, Apr 19, 2011 at 10:48 AM, Guillaume Lelarge
> <guillaume@lelarge.info> wrote:
>> Hi,
>>
>> Le 19/04/2011 10:40, Dave Page a écrit :
>>> [...]
>>> I've tweaked the wording on the message in this patch as I wasn't
>>> entirely happy with the phrasing,
>>
>> Good idea.
>>
>>> but whilst doing so I noticed there
>>> doesn't seem to be any error checking to handle the case where the
>>> extension cannot be installed (for example, because the user is using
>>> a hosting site which hasn't installed the contrib modules).
>>>
>>
>> We first check if the extension is available (query on
>> pg_available_extensions). bUT I should probably add a check on the
>> CREATE EXTENSION query because it may fail (if you're not superuser for
>> example).
>>
>>> There's also no way to permanently suppress the message box, which
>>> could become very annoying.
>>>
>>
>> Yeah, I didn't think about it yesterday. That would be good to add.
>>
>>> Final thought - we also have a guru hint about the admin pack
>>> (instrumentation.html). Shouldn't this patch touch the same places as
>>> that hint (as well as anywhere else it makes sense)?
>>>
>>
>> I've been thinking about it. I was wondering if we should have a
>> specific hint for 8.1 till 9.0 and another one for 9.1.
>>
>> And one other issue: we may be connected to another database than the
>> postgres one. I should probably check that too.
>
> I was thinking more that we shouldn't have a hint in one place, and
> offer to fix it in another. We should have the hint, and offer to fix
> the problem in all places (frmHint can handle "fix" options iirc - for
> example, it can vacuum for you).
>

Got it :)

This patch should fix all the remaining issues, shouldn't it?


--
Guillaume
 http://www.postgresql.fr
 http://dalibo.com

В списке pgadmin-hackers по дате отправления:

Предыдущее
От: Dave Page
Дата:
Сообщение: Re: 1.12.3 built
Следующее
От: Thom Brown
Дата:
Сообщение: Insufficient description in collation mismatch error