O$ - DAM issue - differences between dev and runtime
David Walton
omuser at charlie.housing.admin.nyu.edu
Wed Apr 8 09:45:00 EDT 2009
Fred,
I have built a method to see whether we need to hit the backend when
someone updates. This way I can tell, before hitting the backend,
whether I need to issue the update. I've put this into my table
(super) class $update method, then issue the do default if something
has changed, otherwise, don't bother. I've also written an archive
method which I add to my $update methods to record who changed which
pieces of data and when.
It doesn't really override the built in methods ... rather it enhances
their behavior.
On Apr 7, 2009, at 5:01 PM, Fred Brinkman wrote:
> Bas,
>
> What's your main reason for overriding $insert, $update (and, I
> assume, $delete).
> When your data is prepared in advance, is there any reason to do this?
>
> Just curious,
>
> Fred.
>
>
>
>
>
> Fred Brinkman
> http://www.euromnis.org
>
> **********************************************
> Fred Brinkman Consultancy
> B-1000 Brussels
>
> Tel. +32-474-83 80 80
> Fax +32-2-330 10 31 (on request)
> mailto:fred.brinkman at euromnis.org
> *********************************************
>
>
>
>
>
>
>
>
> Op 7 apr 2009, om 22:50 heeft Bastiaan Olij het volgende geschreven:
>
>> Hmmm,
>>
>> Maybe I mixed them up the wrong way around... It has been awhile and
>> these days I override $insert and $update all together with my own
>> code.
>> Sorry about that,
>>
>> Bastiaan
>>
>> Steve Finger wrote:
>>> Hi,
>>>
>>> Now I'm confused. In my table classes I override $update all the
>>> time
>>> and at the end put
>>>
>>> Do default Returns %flag
>>>
>>> %flag is a local variable and it returns 1 if successful and 0 if
>>> not
>>>
>>> In my only experience with Oracle and SQL Server the Do default
>>> works
>>> the same with the 0 or 1 result.
>>>
>>> I think (but I'm going from memory) there may have been a time it
>>> returned null from an sql statement that ended up being totally
>>> messed
>>> up and I guess the Dam didn't know what to do with it.
>>>
>>> Steve Finger
>>>
>>> Bastiaan Olij wrote:
>>>> Hi David,
>>>>
>>>> As you are overrriding $update, do default is not the correct way
>>>> to
>>>> invoke the standard logic.
>>>>
>>>> Try either doing a "do inherited" or what I find easier/better
>>>> myself:
>>>> do $inherited.$update() returns lvSuccess
>>>>
>>>> Which allows you to parse any additional parameters.
>>>>
>>>> I can't remember off the top of my head when you had to use do
>>>> default.
>>>> I hardly use that one in my code.
>>>>
>>>> Greetz,
>>>>
>>>> Bas
>>>>
>>>> David Walton wrote:
>>>>> Okay... no responses. Let's try again.
>>>>>
>>>>> I have a table method that is called $update. It processes some
>>>>> history, checks that the user has changed some values, then
>>>>> says "Do
>>>>> default" with a return value. The problem is that I get different
>>>>> results from the return value on the do default in runtime than in
>>>>> development.
>>>>>
>>>>> Can someone tell me what I should expect from the return value
>>>>> on a do
>>>>> default? Is it boolean, numeric, etc? It seems to be returning a
>>>>> blank
>>>>> (like '').
>>>>>
>>>>> Thanks.
>>>>>
>>>>
>>>> _____________________________________________________________
>>>> Manage your list subscriptions at http://lists.omnis-dev.com
>>> _____________________________________________________________
>>> Manage your list subscriptions at http://lists.omnis-dev.com
>>>
>>
>>
>> --
>> Kindest Regards,
>>
>> Bastiaan Olij
>> e-mail/MSN: bastiaan at basenlily.nl
>> web: http://www.basenlily.nl
>> Skype: Mux213
>> http://www.linkedin.com/in/bastiaanolij
>>
>> _____________________________________________________________
>> Manage your list subscriptions at http://lists.omnis-dev.com
>
> _____________________________________________________________
> Manage your list subscriptions at http://lists.omnis-dev.com
>
More information about the omnisdev-en
mailing list