On Aug 16, 5:02 pm, Malcolm Tredinnick <[EMAIL PROTECTED]>
wrote:
> The current latest patch fails in a dangerous way, which is why I'm
> against it. Fail safely (so that my data always gets saved when I call
> save()) and it's up for consideration.
My changes latest patch ensures it fails safely.
On Wed, 2007-08-15 at 21:51 -0700, Collin Grady wrote:
> Malcolm Tredinnick said the following:
> > The current latest patch fails in a dangerous way, which is why I'm
> > against it. Fail safely (so that my data always gets saved when I call
> > save()) and it's up for consideration.
>
> Is the
Malcolm Tredinnick said the following:
> The current latest patch fails in a dangerous way, which is why I'm
> against it. Fail safely (so that my data always gets saved when I call
> save()) and it's up for consideration.
Is the failure you mention just the __setattr__ issue, or is there
another
On Wed, 2007-08-15 at 21:09 -0700, Collin Grady wrote:
> Malcolm Tredinnick said the following:
> > See earlier threads for why transparent behaviour is not a very safe
> > idea. If you override save(), for example, it's going to be very easy to
> > end up with things not being saved. This is not
Malcolm Tredinnick said the following:
> See earlier threads for why transparent behaviour is not a very safe
> idea. If you override save(), for example, it's going to be very easy to
> end up with things not being saved. This is not common behaviour, so
> asking somebody to at least pass in an e
On Wed, 2007-08-15 at 18:59 -0600, Norman Harman wrote:
> SmileyChris wrote:
> > Collin has put a lot of effort in to this ticket [1] which is still
> > waiting as a design decision. The latest patch is (apart from lack of
> > docs) ready for check-in in my opinion. Could we have a decision on
> >
Norman Harman said the following:
> Nobody knows me, so my opinion isn't worth much. But, this would be good.
> Better would
> be if it were transparent(properties or some other mechanism to notice when
> fields are set
> and therefore dirty/need updating). I'm fairly surprised something li
SmileyChris wrote:
> Collin has put a lot of effort in to this ticket [1] which is still
> waiting as a design decision. The latest patch is (apart from lack of
> docs) ready for check-in in my opinion. Could we have a decision on
> whether this is worthy?
Nobody knows me, so my opinion isn't wor
Collin has put a lot of effort in to this ticket [1] which is still
waiting as a design decision. The latest patch is (apart from lack of
docs) ready for check-in in my opinion. Could we have a decision on
whether this is worthy?
[1] http://code.djangoproject.com/ticket/4102
--~--~-~--~
