Does it work if you do?
db.nodeRelatedName.relatedID.requires=[IS_IN_DB(db,'node.id','%(id)s',zero=T('select
a
value')),IS_NOT_IN_DB(db(db.nodeRelatedName.nodeID==request.vars.nodeID),'nodeRelatedName.relatedID',error_message=T('combination
node and related name already in database'))]
On Saturday, 13 October 2012 03:19:24 UTC-5, Annet wrote:
>
> In db.py I define the following table:
>
> def set_requirement(nodeRelatedName):
> nodeRelatedName.relatedID.requires=[IS_IN_DB(db,'node.id','%(id)s',zero=T('select
>
> a
> value')),IS_NOT_IN_DB(db(db.nodeRelatedName.nodeID==request.vars.nodeID),'nodeRelatedName.relatedID',error_message=T('combination
>
> node and related name already in database'))]
>
> db.define_table('nodeRelatedName',
> Field('nodeID','reference node',**isnode),
> Field('relatedID','reference
> node',default='',ondelete='CASCADE',notnull=True),
>
> Field('insequence',type='integer',default=0,writable=False,readable=False),
> Field('createdOn',**attributes),
> Field('modifiedOn',update=request.now,**attributes),
> on_define=set_requirement,
> migrate=False)
>
> The problem is that the validator doesn't work, I should prevent users
> from adding the same combination of nodeID and relatedID twice or more, but
> it doesn't. What's the correct syntax to implement this?
>
> Kind regards,
>
> Annet
>
--
