|From:||Guillaume Lelarge <guillaume(at)lelarge(dot)info>|
|To:||Dave Page <dpage(at)postgresql(dot)org>|
|Subject:||Re: Enabling SQL text field in the SQL tab of object dialog|
|Views:||Raw Message | Whole Thread | Download mbox|
Dave Page a écrit :
> On Mon, Jun 16, 2008 at 8:58 PM, Guillaume Lelarge
> <guillaume(at)lelarge(dot)info> wrote:
>> Grmbl... There's a big flaw in my patch. You're right once again.
> Meh. It's happened a couple of times now :-)
>> Problem is, pgAdmin divide the queries to launch in two groups, only for
>> some objects : database and tablespace. The SQL textfield contains the
>> concatenation of the two groups. So we can't let the user change this field
>> if we aren't able to separate the two groups.
>> I think there are two possible ways to deal with this:
>> * the simpler, but confusing, one: add another textfield, put each
>> group in one textfield;
>> * the difficult, but less error prone, one: prevent the user to use the
>> read/write mode when he's on the database's properties (and the same
>> for tablespace).
>> I prefer the latter because it's less confusing for the user (why two SQL
>> textfields?), less error prone (what happens if the user takes SQL textfield
>> 1's contents and put it in SQL textfield 2?).
>> Desactivating the read/write mode for database and tablespace doesn't seem a
>> big issue to me, does it?
> Hmm, I'd forgotten about that. We do it because it allows us to run
> non-transaction safe DDL in separate transactions (which is required
> in 8.3). Actually, this is not a major issue - all that will happen is
> that the server will reject the SQL in 8.3+ if the user includes no
> transaction safe DDL with other commands (or blindly accept it in
> earlier versions). I say we just split the text control on those two
> dialogues, per option 1.
Option 1 seems not very user friendly to me. Are you against option 2?
if yes, why?
|Next Message||Dave Page||2008-06-16 20:40:51||Re: Enabling SQL text field in the SQL tab of object dialog|
|Previous Message||Dave Page||2008-06-16 20:15:55||Re: A fix and a new functionnality for the colour patch|