It looks to me like you're right:

http://tapestry.apache.org/tapestry4/apidocs/src-html/org/apache/tapestry/util/io/DataSqueezerImpl.html

Well, at least you found the problem, and I'll halve the chances of
conflicts in the next release by cutting the prefix down to a single char.

Take care,

Daniel


On 12/18/06, Steve Wells <[EMAIL PROTECTED]> wrote:

Confirm it first, could be the lack of caffeine here, I'd be curious to
see
if you see it in the same way.

On 18/12/06, D&J Gredler <[EMAIL PROTECTED]> wrote:
>
> Hmm... very interesting. I wonder why the squeezer prefix works that
way?
> It
> looks like I probably need to change the BeanForm squeezer prefix to
just
> be
> a single character, in order to minimize conflicts...
>
> On 12/18/06, Steve Wells <[EMAIL PROTECTED]> wrote:
> >
> > hmmm, thanks and good thinking guys not sure it nails it though...
> >
> > The prefix is "P" but I have tried others., BeanForm uses "BP"
> >
> > Barry I think the squeezer type is registered by the getDataClass
> method?
> > Cayenne squeezer:
> > public Class getDataClass() {
> >         return DataObject.class;
> > }
> >
> > BeanForm:
> >     public Class getDataClass() {
> >         return BeanProperty.class;
> >     }
> >
> > ahhh haaaa (20 mins later), when I said I tried others I only tried
"PO"
> > (Persistent Object :) ).  I tried 'q' and 'p' and they both
> > work.  Studying
> > the Tapestry DataSqueezerImpl you see that (well it looks like it to
me)
> > all
> > the characters in the prefix String are registered, using "BP" will
> > reserve
> > both the chars, B and P.  Then my squeezer comes along with "P" and it
> > fails
> > with an error message that is a bit misleading.
> >
> > This is not the behaviour I was expecting.  I was treating
> > SqueezeAdapter.getPrefix() to set a unique String key for a squeezer,
so
> > that "A" is different to "AA".  Now I know but perhaps this trap could
> be
> > removed?
> >
> > Thanks all for your suggestions.
> >
> > Steve
> >
> > On 19/12/06, Barry Books <[EMAIL PROTECTED]> wrote:
> > >
> > > Could it be both sqeezers are useable for the object you are trying
to
> > > sqeeze? I'm not sure how that case is handled but the error looks
> > > similar to the one you get when you specify more that one persist
> > > option.
> > >
> > >
---------------------------------------------------------------------
> > > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > > For additional commands, e-mail: [EMAIL PROTECTED]
> > >
> > >
> >
> >
>
>


Reply via email to