Imho when we think about JSF validation we don't need to mimic existing
features of JSF 2, but we have to see where features are missing.
I.e. BVAL 1.1 will contain method-validation, but JSF completely missed to
integrate it. This is where we should fill the gap.
I am thinking about something like
<h:form>
<f:validateMethod value="#{myBean.updatePassword(myBean.password,
myBean.repeatedPassword)}" />
<h:inputSecret value="#{myBean.password}" />
<h:inputSecret value="#{myBean.repeatedPassword}" />
</h:form>
where updatePassword is validated in the PROCESS_VALIDATION phase (with
the local values of both fields) and ideally instead of calling
MyBean.setPassword and MyBean.setRepeatedPassword in the UPDATE_MODEL
phase it should call MyBean.updatePasswordŠ
Such feature would improve cross-field validation enormously.
WDYT?
Regards,
Arne
P.S.: I just prototyped the feature and it basically worked also I am not
sure about the design (in my prototype f:validateMethod is a JSF
Component...)
Am 25.03.13 22:20 schrieb "Mark Struberg" unter <[email protected]>:
>with JSF-2 this is now almost obsolete, isn't?
>JSF-2 has <f:validateBean> which works perfectly fine.
>
>LieGrue,
>strub
>
>
>
>
>----- Original Message -----
>> From: Thomas Andraschko <[email protected]>
>> To: [email protected]
>> Cc:
>> Sent: Monday, March 25, 2013 5:31 PM
>> Subject: Re: DeltaSpike Roadmap
>>
>> Are you sure? PrimeFaces does not have one. I only know richfaces'
>> graphValidator and it's IMO not that easy and cool as s:validateForm.
>>
>> 2013/3/25 John D. Ament <[email protected]>
>>
>>> I believe all of the major JSF component libraries support some type
>>>of
>>> form validation using BV.
>>>
>>>
>>> On Mon, Mar 25, 2013 at 12:19 PM, Thomas Andraschko <
>>> [email protected]> wrote:
>>>
>>> > Any news about s:validateForm?
>>> > As i said, it's great feature of seam3 and would perfectly fit DS
>> (IMO).
>>> >
>>> > I could also add it to DS if no one have enough time... :)
>>> >
>>> > 2012/12/17 Gerhard Petracek <[email protected]>
>>> >
>>> > > there is a reason why it is just an add-on ;)
>>> > > and yes, if you would do it with the jsf2-api, it would be a bit
>>> easier.
>>> > > but you can use this add-on also for jsf 1.x and some users who
>> don't
>>> > like
>>> > > to use seam are happy with it.
>>> > > however, i just mentioned it as an example to show that cdi
>> isn't
>>> needed
>>> > to
>>> > > implement such a functionality.
>>> > >
>>> > > regards,
>>> > > gerhard
>>> > >
>>> > >
>>> > >
>>> > > 2012/12/17 Lincoln Baxter, III <[email protected]>
>>> > >
>>> > > > Also, note that the use-cases for <s:validateForm> and
>> the latter
>>> > example
>>> > > > are different in their usage. IMO, <s:validateForm> is
>> much more
>>> > > > declarative and simple than the example provided from
>> extval.
>>> > > >
>>> > > > Just write a validator and attach it like you would normal
>> to a
>>> > component
>>> > > > in the view, wherever you want to use it. You can also
>> re-use
>>> > > > form-validators using this method, instead of using some (in
>> my
>>> > opinion)
>>> > > > strange view-filtering logic in the FormValidator
>> annotation.
>>> > > >
>>> > > > <s:validateForm> is/was one of the most well recieved
>> feature of Seam
>>> > > > Faces, and I think it deserves serious consideration. Can
>> you do form
>>> > > > validation other ways? absolutely, but this way is pretty
>> compelling
>>> > > based
>>> > > > on the feedback we got. It also ties in very normally with
>> the way in
>>> > > which
>>> > > > JSF handles validation and components, which is not
>> something that I
>>> > can
>>> > > > speak about with regard to extval, but I can say that seam
>> faces does
>>> > > this
>>> > > > nicely.
>>> > > >
>>> > > > ~Lincoln
>>> > > >
>>> > > >
>>> > > > On Mon, Dec 17, 2012 at 4:37 PM, Thomas Andraschko <
>>> > > > [email protected]> wrote:
>>> > > >
>>> > > > > Please do not mix both :)
>>> > > > > We had the discussion about commons bv constrains.
>> s:validateForm
>>> is
>>> > > > > completely different compared to bv constraints.
>>> > > > > s:validateForm does not fit the project for common bv
>> constraints.
>>> > > > >
>>> > > > > 2012/12/17 Karl Kildén <[email protected]>
>>> > > > >
>>> > > > > > Looks really nice I understand finally :-)
>>> > > > > >
>>> > > > > > In retrospect I don't know why I didn't
>> check the docs myself,
>>> > > probably
>>> > > > > > because I read most of the discussion on the phone
>> to begin with.
>>> > > > > >
>>> > > > > > I then agree with the many suggestions regarding
>> doing it as a
>>> > > parallel
>>> > > > > > project. Perhaps that module could have a sub
>> module for CDI
>>> users
>>> > > > that
>>> > > > > > included producers and such.
>>> > > > > >
>>> > > > > > Best regards / Karl
>>> > > > > >
>>> > > > > >
>>> > > > > >
>>> > > > > >
>>> > > > > > 2012/12/17 Gerhard Petracek
>> <[email protected]>
>>> > > > > >
>>> > > > > > > hi karl,
>>> > > > > > >
>>> > > > > > > it's implemented via a producer - see
>> [1].
>>> > > > > > > but you don't really need cdi for it -
>> see e.g. [2].
>>> > > > > > >
>>> > > > > > > regards,
>>> > > > > > > gerhard
>>> > > > > > >
>>> > > > > > > [1] http://s.apache.org/ury
>>> > > > > > > [2]
>>> > > > > > >
>>> > > > > > >
>>> > > > > >
>>> > > > >
>>> > > >
>>> > >
>>> >
>>>
>>
>>http://os890.blogspot.co.at/2010/06/multi-field-form-validation-with-jsr.
>>html
>>> > > > > > >
>>> > > > > > >
>>> > > > > > >
>>> > > > > > > 2012/12/17 Thomas Andraschko
>> <[email protected]>
>>> > > > > > >
>>> > > > > > > > As far as i understand, validateAll just
>> validates all
>>> > components
>>> > > > but
>>> > > > > > you
>>> > > > > > > > can't use custom logic.
>>> > > > > > > >
>>> > > > > > > > Checkout the seam docu here:
>>> > > > > > > >
>>> > > > > > > >
>>> > > > > > >
>>> > > > > >
>>> > > > >
>>> > > >
>>> > >
>>> >
>>>
>>
>>http://docs.jboss.org/seam/3/faces/latest/reference/en-US/html/components
>>.html
>>> > > > > > > >
>>> > > > > > > > 2012/12/17 Karl Kildén
>> <[email protected]>
>>> > > > > > > >
>>> > > > > > > > > Hrmm still not sure I understand.
>> First off didn't Pete
>>> just
>>> > > say
>>> > > > it
>>> > > > > > was
>>> > > > > > > > > just a JSF-validator? It also
>> sounds and reads just like
>>> > > > > validateAll
>>> > > > > > > from
>>> > > > > > > > > omnifaces?
>>> > > > > > > > >
>>> > > > > > > > > What is lacking that you would like
>> to see solved by CDI?
>>> And
>>> > > is
>>> > > > > it a
>>> > > > > > > > > natural fit for CDI extensions
>> rather then BV / JSF
>>> > extensions?
>>> > > > > > > > >
>>> > > > > > > > > Best regards / Karl
>>> > > > > > > > >
>>> > > > > > > > >
>>> > > > > > > > >
>>> > > > > > > > > 2012/12/17 Thomas Andraschko
>> <[email protected]>
>>> > > > > > > > >
>>> > > > > > > > > > Hi,
>>> > > > > > > > > >
>>> > > > > > > > > > they are nice but i can't
>> see a replacement for
>>> > > s:validateForm.
>>> > > > > > > > > > IMO s:validateForm is really a
>> simple solution for
>>> > > cross-field
>>> > > > > > > > > validation,
>>> > > > > > > > > > which is CDI based.
>>> > > > > > > > > >
>>> > > > > > > > > > Regards,
>>> > > > > > > > > > THomas
>>> > > > > > > > > >
>>> > > > > > > > > > 2012/12/17 Karl Kildén
>> <[email protected]>
>>> > > > > > > > > >
>>> > > > > > > > > > > Well these are the
>> validators from omnifaces as per my
>>> > > > previous
>>> > > > > > > > > > > recommendation:
>>> > > > > > > > > > >
>>> > > > > > > > > > >
>>> > > > > > > > > > > -
>> RequiredCheckboxValidator<
>>> > > > > > > > > > >
>>> > > > > > > > > >
>>> > > > > > > > >
>>> > > > > > > >
>>> > > > > > >
>>> > > > > >
>>> > > > >
>>> > > >
>>> > >
>>> >
>>>
>>
>>https://showcase-omnifaces.rhcloud.com/showcase/validators/RequiredCheckb
>>oxValidator.xhtml
>>> > > > > > > > > > > >
>>> > > > > > > > > > > - validateAll<
>>> > > > > > > > > > >
>>> > > > > > > > > >
>>> > > > > > > > >
>>> > > > > > > >
>>> > > > > > >
>>> > > > > >
>>> > > > >
>>> > > >
>>> > >
>>> >
>>>
>>
>>https://showcase-omnifaces.rhcloud.com/showcase/validators/validateAll.xh
>>tml
>>> > > > > > > > > > > >
>>> > > > > > > > > > > -
>> validateAllOrNone<
>>> > > > > > > > > > >
>>> > > > > > > > > >
>>> > > > > > > > >
>>> > > > > > > >
>>> > > > > > >
>>> > > > > >
>>> > > > >
>>> > > >
>>> > >
>>> >
>>>
>>
>>https://showcase-omnifaces.rhcloud.com/showcase/validators/validateAllOrN
>>one.xhtml
>>> > > > > > > > > > > >
>>> > > > > > > > > > > - validateEqual<
>>> > > > > > > > > > >
>>> > > > > > > > > >
>>> > > > > > > > >
>>> > > > > > > >
>>> > > > > > >
>>> > > > > >
>>> > > > >
>>> > > >
>>> > >
>>> >
>>>
>>
>>https://showcase-omnifaces.rhcloud.com/showcase/validators/validateEqual.
>>xhtml
>>> > > > > > > > > > > >
>>> > > > > > > > > > > - validateOne<
>>> > > > > > > > > > >
>>> > > > > > > > > >
>>> > > > > > > > >
>>> > > > > > > >
>>> > > > > > >
>>> > > > > >
>>> > > > >
>>> > > >
>>> > >
>>> >
>>>
>>
>>https://showcase-omnifaces.rhcloud.com/showcase/validators/validateOne.xh
>>tml
>>> > > > > > > > > > > >
>>> > > > > > > > > > > -
>> validateOneOrMore<
>>> > > > > > > > > > >
>>> > > > > > > > > >
>>> > > > > > > > >
>>> > > > > > > >
>>> > > > > > >
>>> > > > > >
>>> > > > >
>>> > > >
>>> > >
>>> >
>>>
>>
>>https://showcase-omnifaces.rhcloud.com/showcase/validators/validateOneOrM
>>ore.xhtml
>>> > > > > > > > > > > >
>>> > > > > > > > > > > -
>> validateOneOrNone<
>>> > > > > > > > > > >
>>> > > > > > > > > >
>>> > > > > > > > >
>>> > > > > > > >
>>> > > > > > >
>>> > > > > >
>>> > > > >
>>> > > >
>>> > >
>>> >
>>>
>>
>>https://showcase-omnifaces.rhcloud.com/showcase/validators/validateOneOrN
>>one.xhtml
>>> > > > > > > > > > > >
>>> > > > > > > > > > > - validateOrder<
>>> > > > > > > > > > >
>>> > > > > > > > > >
>>> > > > > > > > >
>>> > > > > > > >
>>> > > > > > >
>>> > > > > >
>>> > > > >
>>> > > >
>>> > >
>>> >
>>>
>>
>>https://showcase-omnifaces.rhcloud.com/showcase/validators/validateOrder.
>>xhtml
>>> > > > > > > > > > > >
>>> > > > > > > > > > > - validateUnique
>>> > > > > > > > > > >
>>> > > > > > > > > > >
>>> > > > > > > > > > > <
>>> > > > > > > > > > >
>>> > > > > > > > > >
>>> > > > > > > > >
>>> > > > > > > >
>>> > > > > > >
>>> > > > > >
>>> > > > >
>>> > > >
>>> > >
>>> >
>>>
>>
>>https://showcase-omnifaces.rhcloud.com/showcase/validators/validateUnique
>>.xhtml
>>> > > > > > > > > > > >Best
>>> > > > > > > > > > > regards / Karl
>>> > > > > > > > > > >
>>> > > > > > > > > > >
>>> > > > > > > > > > >
>>> > > > > > > > > > > 2012/12/17 Thomas
>> Andraschko <
>>> > [email protected]>
>>> > > > > > > > > > >
>>> > > > > > > > > > > > PrimeFaces
>> doesn't have such a component/utility.
>>> > > > > > > > > > > > Also AFAICS it's
>> based on CDI, so IMO DeltaSpike is a
>>> > > good
>>> > > > > > place
>>> > > > > > > > for
>>> > > > > > > > > > it.
>>> > > > > > > > > > > >
>>> > > > > > > > > > > > 2012/12/17 Jason
>> Porter <[email protected]>
>>> > > > > > > > > > > >
>>> > > > > > > > > > > > > On Mon, Dec 17,
>> 2012 at 4:54 AM, Pete Muir <
>>> > > > > [email protected]
>>> > > > > > >
>>> > > > > > > > > wrote:
>>> > > > > > > > > > > > >
>>> > > > > > > > > > > > > >
>>> > > > > > > > > > > > > > On 17 Dec
>> 2012, at 08:55, Gerhard Petracek wrote:
>>> > > > > > > > > > > > > >
>>> > > > > > > > > > > > > > > hi
>> karl,
>>> > > > > > > > > > > > > > >
>>> > > > > > > > > > > > > > > #1
>> apache myfaces (extval) doesn't implement
>>> jsr
>>> > > 303
>>> > > > > > (e.g.
>>> > > > > > > > > apache
>>> > > > > > > > > > > > > > > bval
>> implements it)
>>> > > > > > > > > > > > > > > #2
>> there is no agreement that ds is only
>>> backend
>>> > > > > oriented
>>> > > > > > > > > > > > > > >
>>> > > > > > > > > > > > > > >
>> regards,
>>> > > > > > > > > > > > > > >
>> gerhard
>>> > > > > > > > > > > > > > >
>>> > > > > > > > > > > > > > >
>>> > > > > > > > > > > > > > >
>>> > > > > > > > > > > > > > >
>> 2012/12/17 Karl Kildén <[email protected]>
>>> > > > > > > > > > > > > > >
>>> > > > > > > > > > > > > > >>
>> Hi Thomas, fellow user here.
>>> > > > > > > > > > > > > > >>
>>> > > > > > > > > > > > > > >>
>> BV is as you may know specified in JSR-303. It
>>> > has
>>> > > > > great
>>> > > > > > > > > > > > > >
>> implementations by
>>> > > > > > > > > > > > > > >>
>> myfaces and hibernate that you can find
>>> easily.
>>> > > Thus
>>> > > > > > this
>>> > > > > > > > will
>>> > > > > > > > > > not
>>> > > > > > > > > > > > be
>>> > > > > > > > > > > > > > >>
>> implemented in deltaspike afaik. However
>>> > providing
>>> > > > > some
>>> > > > > > > > common
>>> > > > > > > > > > > > > >
>> constraints
>>> > > > > > > > > > > > > > >>
>> beyond the specification sounds like a good
>>> > idea.
>>> > > > Note
>>> > > > > > > that
>>> > > > > > > > in
>>> > > > > > > > > > the
>>> > > > > > > > > > > > > mean
>>> > > > > > > > > > > > > > >>
>> while @Email is almost trivial to implement
>>> > > yourself
>>> > > > > if
>>> > > > > > > you
>>> > > > > > > > > > search
>>> > > > > > > > > > > > it
>>> > > > > > > > > > > > > a
>>> > > > > > > > > > > > > > >>
>> little. JSR-303 is highly extensible just like
>>> > > CDI.
>>> > > > > The
>>> > > > > > > > > > > > implementation
>>> > > > > > > > > > > > > > you
>>> > > > > > > > > > > > > > >>
>> choose might have it as an extra already.
>>> > > > > > > > > > > > > > >>
>>> > > > > > > > > > > > > > >>
>> For frontend extras I can't say I see an
>>> obvious
>>> > > > home
>>> > > > > in
>>> > > > > > > > > > > deltaspike
>>> > > > > > > > > > > > > > >>
>> (someone correct me if I am wrong about this).
>>> > > But I
>>> > > > > > think
>>> > > > > > > > for
>>> > > > > > > > > > now
>>> > > > > > > > > > > > > > >>
>> Deltaspike is more backend oriented. I
>>> recommend
>>> > > > > having
>>> > > > > > a
>>> > > > > > > > look
>>> > > > > > > > > > at
>>> > > > > > > > > > > > > > omnifaces
>>> > > > > > > > > > > > > > >>
>> that already provide several great custom
>>> > > > > jsf-validators
>>> > > > > > > and
>>> > > > > > > > > has
>>> > > > > > > > > > > an
>>> > > > > > > > > > > > > open
>>> > > > > > > > > > > > > > >>
>> request for users to ask for additional ones.
>>> > > > > > > > > > > > > > >>
>>> > > > > > > > > > > > > > >>
>> best regards / Karl
>>> > > > > > > > > > > > > > >>
>>> > > > > > > > > > > > > > >>
>>> > > > > > > > > > > > > > >>
>>> > > > > > > > > > > > > > >>
>>> > > > > > > > > > > > > > >>
>> 2012/12/16 Thomas Andraschko <
>>> > > > > > [email protected]
>>> > > > > > > >
>>> > > > > > > > > > > > > > >>
>>> > > > > > > > > > > > > >
>>>>> Hi,
>>> > > > > > > > > > > > > >
>>>>>
>>> > > > > > > > > > > > > >
>>>>> is there a roadmap available for all upcoming
>>> > > > > features?
>>> > > > > > > > > > > > > >
>>>>>
>>> > > > > > > > > > > > > >
>>>>> Will there also be a BV module in the future?
>>> > > Will
>>> > > > it
>>> > > > > > > also
>>> > > > > > > > > > > include
>>> > > > > > > > > > > > > some
>>> > > > > > > > > > > > > >
>>>>> commons constraints like Email?
>>> > > > > > > > > > > > > >
>>> > > > > > > > > > > > > > Hmm, I
>> think this could be a good thing for a
>>> > > parallel
>>> > > > > > > project
>>> > > > > > > > to
>>> > > > > > > > > > > > > > DeltaSpike
>> actually - a library of BV constraints
>>> > not
>>> > > > > > branded
>>> > > > > > > > by
>>> > > > > > > > > a
>>> > > > > > > > > > > > > specific
>>> > > > > > > > > > > > > > impl.
>> I've cc'd Emmanuel, spec lead for BV, to
>>> see
>>> > if
>>> > > > he
>>> > > > > > has
>>> > > > > > > > any
>>> > > > > > > > > > > plans
>>> > > > > > > > > > > > > for
>>> > > > > > > > > > > > > > such a
>> thing.
>>> > > > > > > > > > > > > >
>>> > > > > > > > > > > > > >
>>>>>
>>> > > > > > > > > > > > > >
>>>>> What about the Seam s:validateForm? IMO it's
>>> a
>>> > > > simple
>>> > > > > > way
>>> > > > > > > > for
>>> > > > > > > > > > > cross
>>> > > > > > > > > > > > > > field
>>> > > > > > > > > > > > > >
>>>>> validation. Will this be added in future
>>> > > releases?
>>> > > > > > > > > > > > > >
>>> > > > > > > > > > > > > > This is
>> not actually a BV validator, but a JSF
>>> > > > > validator. I
>>> > > > > > > > know
>>> > > > > > > > > > Mark
>>> > > > > > > > > > > > is
>>> > > > > > > > > > > > > > working on
>> JSF stuff for DeltaSpike atm, perhaps
>>> > this
>>> > > > is
>>> > > > > > > > > something
>>> > > > > > > > > > > > Jason
>>> > > > > > > > > > > > > > could add
>> to that.
>>> > > > > > > > > > > > > >
>>> > > > > > > > > > > > >
>>> > > > > > > > > > > > > Is this
>> something that belongs in DeltaSpike, or
>>> > should
>>> > > > it
>>> > > > > be
>>> > > > > > > in
>>> > > > > > > > a
>>> > > > > > > > > > JSF
>>> > > > > > > > > > > > > component
>> library? I know RichFaces has something
>>> > very
>>> > > > > > > similar, I
>>> > > > > > > > > > > believe
>>> > > > > > > > > > > > > PrimeFaces does
>> as well (haven't looked for a
>>> while)
>>> > > and
>>> > > > I
>>> > > > > > have
>>> > > > > > > > no
>>> > > > > > > > > > idea
>>> > > > > > > > > > > > > about other
>> component libraries.
>>> > > > > > > > > > > > >
>>> > > > > > > > > > > > >
>>> > > > > > > > > > > > > >
>>>>>
>>> > > > > > > > > > > > > >
>>>>> How can i contribue to DeltaSpike? I could do
>>> > > some
>>> > > > > > > smaller
>>> > > > > > > > > > tasks
>>> > > > > > > > > > > > > > without
>>> > > > > > > > > > > > > >
>>>>> problems.
>>> > > > > > > > > > > > > >
>>>>> I also implemented constraints like
>>> > > > > > > > >
>> EqualsExpression("#{...}").
>>> > > > > > > > > > > > Maybe
>>> > > > > > > > > > > > > > >>
>> this
>>> > > > > > > > > > > > > >
>>>>> could be useful for other users, too.
>>> > > > > > > > > > > > > >
>>> > > > > > > > > > > > > > This page
>> should help -
>>> > > > > > > > > > > > > >
>>> > > > > > >
>> http://deltaspike.staging.apache.org/deltaspike/community.html
>>> > > > > > > > > > > > > >
>>> > > > > > > > > > > > > >
>>>>>
>>> > > > > > > > > > > > > >
>>>>> Best regards,
>>> > > > > > > > > > > > > >
>>>>> Thomas
>>> > > > > > > > > > > > > >
>>>>>
>>> > > > > > > > > > > > > > >>
>>> > > > > > > > > > > > > >
>>> > > > > > > > > > > > > >
>>> > > > > > > > > > > > >
>>> > > > > > > > > > > > >
>>> > > > > > > > > > > > > --
>>> > > > > > > > > > > > > Jason Porter
>>> > > > > > > > > > > > >
>> http://lightguard-jp.blogspot.com
>>> > > > > > > > > > > > >
>> http://twitter.com/lightguardjp
>>> > > > > > > > > > > > >
>>> > > > > > > > > > > > > Software
>> Engineer
>>> > > > > > > > > > > > > Open Source
>> Advocate
>>> > > > > > > > > > > > >
>>> > > > > > > > > > > > > PGP key id:
>> 926CCFF5
>>> > > > > > > > > > > > > PGP key
>> available at: keyserver.net, pgp.mit.edu
>>> > > > > > > > > > > > >
>>> > > > > > > > > > > >
>>> > > > > > > > > > >
>>> > > > > > > > > >
>>> > > > > > > > >
>>> > > > > > > >
>>> > > > > > >
>>> > > > > >
>>> > > > >
>>> > > >
>>> > > >
>>> > > >
>>> > > > --
>>> > > > Lincoln Baxter, III
>>> > > > http://ocpsoft.org
>>> > > > "Simpler is better."
>>> > > >
>>> > >
>>> >
>>>
>>
