my recommendation has always been to let wicket generate the ids.

the css argument is invalid imho because you can just as easily use a class
instead of id

the javascript does carry some weight. if the javascript is inlined/static
and in the page then yeah, it sucks that wicket would change the attribute,
that is why we have a setmarkupid() that lets you override it. if you use it
it is then up to you to make sure the id you set is unique.

personally i keep my js snippets in .js files and use texttemplate with
variable replacement to stick them into head. that way all my needed ids are
just variables whose value i collect before outputting the javascript.

-igor


On 9/6/07, Johan Compagner <[EMAIL PROTECTED]> wrote:
>
> the wicketid doesn't have to be unique only if you are setting that id
> right
> into the markupid yourself.
> if not then wicket makes them unique (if we didn't do that THEN wicket ids
> should be unique)
>
> The 2.0 way we just can't do because we don't know it early enough now
> only
> in onBeforeRender
> igor did try some things out i believe
>
> But what would be nice if we try to get it from the html. But if you call
> it
> in your code to soon then we generate one.
> But even if we take it from the html then still if the component is reused
> we have a problem because it is not unique
>
> johan
>
>
> On 9/6/07, Sam Hough <[EMAIL PROTECTED]> wrote:
> >
> >
> > Guess the worst thing is that Wicket ids now have to be unique within
> the
> > whole page not just the parent component...
> >
> > The behaviour they had in 2.0 does seem better.
> > --
> > View this message in context:
> >
> http://www.nabble.com/getMarkupId-doesn%27t-return-the-id-from-the-markup-tf3972925.html#a12525262
> > Sent from the Wicket - Dev mailing list archive at Nabble.com.
> >
> >
>

Reply via email to