Validators in a model's validator_list (m-r)

Christian Schneider wrote:
>; On 4/25/06, Michael Radziej <noris.de">mirnoris.de&gt; wrote:
>;>
>&gt; Christian Schneider wrote:
>;>> Michael,
&gt;>>
>>&gt; I'm still using a generic view. I played with some custom functions
>> before
>;>> and they were called so I'd hoped that it would work with the validator
>>>; objects as well.
>;> I looked in the source. Aha! First parameter to RequiredIfOtherField is
>>; the *name* of the field (a string, not a field).
&gt;>
>> Please try and report ...
>
&gt;
> nothing's happening, :-(
>
> Even if it worked would validation stop if the referenced field didn't have
> the required value? That is: Field 1 is required when field 2 has value "x&quot;
> and field 2's value is "y&quot;. Field 1 is blank which it should be allowed to
> but would validation stop after evaluating the RequiredIfOtherField
> validator runs (if it even runs first)?

I don't know what you mean with "stop". I think each validator will be
executed, but they will take into account whether record_type has the
right value.

Anyway, I personally give up :-(

You can try to find out what's happening:

- create your own validator that always raises a ValidationError (see in
the sources for core.validators to see how)
- add this to your validator list
- do you get a validation message?
&nbsp; - if not, do something like "raise Blablabla&quot; which should immediately
result in a backtrace when called. What's happening now?
&nbsp; - else, change the source of core.validators in Django; raise
Blablabla in the __call__() method. Do you get the error message?


You get the picture ...

Hope that helps a little bit.

I created a ticket to add documentation for this module (#1680)

Christian Schneider wrote:
>; On 4/25/06, Michael Radziej <noris.de">mirnoris.de&gt; wrote:
>;>
>&gt; Christian Schneider wrote:
>;>> Michael,
&gt;>>
>>&gt; I'm still using a generic view. I played with some custom functions
>> before
>;>> and they were called so I'd hoped that it would work with the validator
>>>; objects as well.
>;> I looked in the source. Aha! First parameter to RequiredIfOtherField is
>>; the *name* of the field (a string, not a field).
&gt;>
>> Please try and report ...
>
&gt;
> nothing's happening, :-(
>
> Even if it worked would validation stop if the referenced field didn't have
> the required value? That is: Field 1 is required when field 2 has value "x&quot;
> and field 2's value is "y&quot;. Field 1 is blank which it should be allowed to
> but would validation stop after evaluating the RequiredIfOtherField
> validator runs (if it even runs first)?

I don't know what you mean with "stop". I think each validator will be
executed, but they will take into account whether record_type has the
right value.

Anyway, I personally give up :-(

You can try to find out what's happening:

- create your own validator that always raises a ValidationError (see in
the sources for core.validators to see how)
- add this to your validator list
- do you get a validation message?
&nbsp; - if not, do something like "raise Blablabla&quot; which should immediately
result in a backtrace when called. What's happening now?
&nbsp; - else, change the source of core.validators in Django; raise
Blablabla in the __call__() method. Do you get the error message?


You get the picture ...

Hope that helps a little bit.

I created a ticket to add documentation for this module (#1680)