I cannot argue. You make a great case. But there must be leniance
for those who want to generate other attributes. I guess I don't
have the answer yet. :-)

--- Martin Cooper <[EMAIL PROTECTED]> wrote:

> On 3/26/06, Paul Benedict <[EMAIL PROTECTED]> wrote:
> >
> > I am late on this conversation, but I want to offer a suggestion.
> > Some tag libraries which render HTML elements allow you to
> > pass-through other attributes. In JSP 2.1 (maybe 2.0 too), there is
> > an implicit attributes object for any attributes which were not
> > explicitly defined. In the ValueList tag library, they include
> > an <attribute> parameter which allows people to define any other
> > parameters they would like to pass in:
> >
> > http://valuelist.sourceforge.net/tlddoc/index.html
> >
> > If you wanted to do something similar:
> > <struts:textbox ...>
> >   <struts:attribute name="autocomplete" value="autocomplete"/>
> > </struts:textbox>
> >
> > I agree that only HTML attributes should be supported, but we should
> > also allow people to pass in other attributes in a way that we don't
> > have to support. I vote for this kind of approach.
> 
> 
> Something like this has been suggested before. The problem I see with it,
> other than the fundamental one of allowing Struts to generate non-standard
> HTML, is that people are then going to come crying to us when it doesn't
> work for them when they're using non-standard attributes, even if we claim
> we don't support that scenario. And, to be honest, who would use such a
> mechanism for standard attributes when they're already supported by the
> tags? ;-)
> 
> --
> Martin Cooper
> 
> 
> Paul
> >
> > --- Mark Lowe <[EMAIL PROTECTED]> wrote:
> >
> > > I just went to submit the a patch, and did a search for autocomplete
> > > before I added a new bug and fund 5 or so entries came up.. The last
> > > ones were 2004, all the comments basically said "autocomplete isn't
> > > part of the html spec so struts html taglib wont do it"..
> > >
> > > While I know the html spec hasn't changed in that it still doesn't
> > > include autocomplete, not supporting it basically makes struts harder
> > > to use. Some one with some html knowledge gets struts, creates a form
> > > .. "Drat s/he thinks that pesky autocomplete is filling in my forms
> > > and it shouldn't be", s/he wants to deactive autocomplete. S/he goes
> > > to the struts list and asks "when i use the autocomplete attribute i
> > > get an error saying its not supported how do i fix this?". Someone
> > > trying to be helpful says "oh you subclass the tag and edit the tld".
> > > Someone who's a pretty good page builder is now potentially completely
> > > disempowered, something relatively simple has been made more complex,
> > > perhaps outside his/her specialist field. Also the next time s/he
> > > needs to upgrade struts there are more hoops to jump through, if
> > > nothing else ensuring the tld is updated, sure 1 additional step, but
> > > an additional step all the same.
> > >
> > > While I do understand the motivations for the party line on this, I
> > > think its cutting off one's nose to spite the face. But thats just my
> > > opinion.
> > >
> > > Is it worth me submiting the patches?
> > >
> > > Mark
> > >
> > > On 3/24/06, Mark Lowe <[EMAIL PROTECTED]> wrote:
> > > > On 3/24/06, Niall Pemberton <[EMAIL PROTECTED]> wrote:
> > > > > With maven 1 you run maven jar:install on the "action" sub-project -
> > > > > that should create the jar and install it in your local maven
> > > > > repository. Then you can go on and build the taglib sub-project and
> > it
> > > > > should find the action jar.
> > > > >
> > > > > Theres been a long standing policy of "only HTML 4 attributes" will
> > be
> > > > > allowed in the tags and I would be surprised if this change wasn't
> > > > > vetoed on this basis,
> > > >
> > > > Yeah I know, but I thought I'd give it a go. Using
> > > > renderOtherAttributes seems an okay option, but users using struts out
> > > > the box might find the idea of editing the tag files a pain in the
> > > > backside. I can see why sticking to the spec is the party line, but
> > > > autocomplete causes problems for folk. I guess its wait and see what
> > > > you guys think.
> > > >
> > > >  even if you find someone willing to commit it.
> > > > >
> > > > > Niall
> > > > >
> > > > > On 3/24/06, Mark Lowe <[EMAIL PROTECTED]> wrote:
> > > > > > CAn anyone give me a crash course on how I can test my changes?
> > I've
> > > > > > made the modifications I need to make, and added the unit tests..
> > The
> > > > > > maven 2 build is complaining it cant download
> > > > > >
> > > > > >
> > http://cvs.apache.org/repository/struts/jars/struts-action-1.3.1-SNAPSHOT.jar
> > > > > >
> > > > > > I've been lazy, cd taglibs and then tried mvn test.. (yes
> > > > > > taglib/build.properties is done.) should I be using maven 1? I
> > know I
> > > > > > can find the answers if i dig into the build, but if someone can
> > fast
> > > > > > track me i'd appeciate it.
> > > > > >
> > > > > > Mark
> > > > > >
> > > > > > On 3/24/06, Matthias Wessendorf <[EMAIL PROTECTED]>
> > wrote:
> > > > > > > > I know that autocomplete is a non standard attribute so might
> > > > > > > > not be accepted as a patch on the taglib. But if it only
> > > > > > > > rendered if specified then it wouldn't force users to use non
> > > > > > > > standard stuff.
> > > > > > >
> > > > > > >
> > > > > > > JSF 1.2 goes the same route. The autocomplete attribute has been
> > > > > > > added to <h:inputText /> component by Ed Burns.
> > > > > > >
> > > > > > > so I am +1 on *patching* Struts' taglib
> > > > > > >
> > > > > > > -Matthias
> > > > > > >
> > > > > > > > If such a patch is likely to make it into future versions
> > > > > > > > I'll get on it.
> > > > > > > >
> > > > > > > > Mark
> > > > > > > >
> > > > > > > >
> > ---------------------------------------------------------------------
> > > > > > > > To unsubscribe, e-mail: [EMAIL PROTECTED] For
> > > > > > > > additional commands, e-mail: [EMAIL PROTECTED]
> > > > > > > >
> > > > > > > >
> > > > > > >
> > > > > > >
> > ---------------------------------------------------------------------
> > > > > > > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > > > > > > For additional commands, e-mail: [EMAIL PROTECTED]
> > > > > > >
> > > > > > >
> > > > > >
> > > > > >
> > ---------------------------------------------------------------------
> > > > > > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > > > > > For additional commands, e-mail: [EMAIL PROTECTED]
> > > > > >
> > > > > >
> > > > >
> > > > >
> > ---------------------------------------------------------------------
> > > > > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > > > > For additional commands, e-mail: [EMAIL PROTECTED]
> > > > >
> > > > >
> > > >
> > >
> > > ---------------------------------------------------------------------
> > > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > > For additional commands, e-mail: [EMAIL PROTECTED]
> > >
> > >
> >
> >
> > __________________________________________________
> > Do You Yahoo!?
> > Tired of spam?  Yahoo! Mail has the best spam protection around
> > http://mail.yahoo.com
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: [EMAIL PROTECTED]
> > For additional commands, e-mail: [EMAIL PROTECTED]
> >
> >
> 


__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to