On Apr 2, 6:57 pm, Clemens Oertel <clemens.oer...@gmail.com> wrote:
> One follow up question, relating to the contexts I mentioned earlier:
> Following my approach mentioned below, let's assume I have a template
> for embedding. This template contains the HTML code for a tabular
> list of data entities. Depending on where the template is embedded,
> some of the entities' fields should appear as links: There are 2
> linkable reference fields A and B, and depending on the context,
> either A, B, or A and B should be rendered as links to the referenced
> entity.
>
> This logic obviously can't go into the template. Also, the model's
> toForm method is unsuitable (having the model referring to a
> RequestVar seems evil to me). No no, this should go into the snippet,
> where such logic belongs. But this would mean having to go through the
> templates XML tree before calling Site.toForm(site, form), enclosing
> all affected liftf:field elements with the respective links. This
> sounds very, very wrong.
>
> Thus, from my current understanding, this context-sensitive rendering
> of fields is best done using the explicit lift-tag/bind approach. So
> I'll probably have to have explicit bind calls anyways, and won't be
> able to use my initial approach of having the mapper class (or it's
> companion object) fill out the forms itself.
>
> I'd be happy if someone could prove me wrong ...
I don't necessarily see this in terms of right and wrong. Well your
embeddable template could use a snippet and you can potentially use
chooseTemplate to kind of shape the form template ... then you can
call toForm in this snippet hence passing only the appropriate form
template and not a huge template containing things that do not relate
with your form per se.
>
> Best,
> Clemens
>
> PS: Exercise for the reader: Instead of selective linking, have only
> either A or B be displayed, depending on the embedding context.
>
> On 31-Mar-09, at 3:59 PM, marius d. wrote:
>
>
>
>
>
> >> ---- Outer Template ----
> >> <lift:surround with="default" at="content">
> >> <h2>Create a New Site</h2>
>
> >> <lift:snippet type="SiteOps.add" form="post" eager_eval="true">
> >> <table>
> >> <tfoot>
> >> <tr>
> >> <td colspan="3"><site:submit>Submit</site:submit></td>
> >> </tr>
> >> </tfoot>
> >> <lift:embed what="/sites/_site" />
> >> </table>
> >> </lift:snippet>
> >> </lift:surround>
> >> ---- End Outer Template ----
>
> >> ---- Embedded Template ----
> >> <tbody>
> >> <tr>
> >> <td><liftf:field_label name="name">Name</liftf:field_label>:</
> >> td>
> >> <td><liftf:field name="name">A Hospital Site</liftf:field></td>
> >> <td><liftf:field_msg name="name" /></td>
> >> </tr>
> >> </tbody>
> >> ---- End Embedded Template ----
>
> >> ---- SiteOps ----
> >> def add(form: NodeSeq): NodeSeq = {
> >> val invokedAs = S.invokedAs
> >> val site = Site.create
>
> >> def newSite(form: NodeSeq): NodeSeq = {
> >> def saveMe(): Unit = {
> >> site.validate match {
> >> case Nil => site.save ; S.notice("Added " + site.name);
> >> S.redirectTo("/sites/")
> >> case xs => S.error(xs) ; S.mapSnippet(invokedAs, newSite)
> >> }
> >> }
>
> >> bind("site", Site.toForm(site, form), "submit" ->
> >> submit("Save", saveMe))
> >> }
>
> >> newSite(form)
> >> }
> >> ---- End SiteOps ----
>
> >> The Site.toForm function is pretty much the same as found in
> >> MetaRecord. It will eventually call each fields' toForm function
> >> (which are the original Lift 1.0 mapper versions, no changes there).
>
> >> Am I really doing things outside the rendering pipeline (this is
> >> not a
> >> rhetorical question)?
>
> > Nope sorry ... I misunderstood your case. By bad entirely.
>
> > It all seems to work, incl. validation.
> >> Also, the eagerly evaluated template, which will be passed to toForm,
> >> contains only field tags, no other lift tags.
>
> >> Again, imagine a few dozen different entities, each with a ton of
> >> fields. I'm sure you'll understand that I'm hesitant to bind each
> >> field manually (once for the list page, once for the display page,
> >> once for the add page, and again once for the edit page, even though
> >> the last 2 can surely be combined into one bind call in the code).
> >> I obviously have to declare each field in the model, and refer to it
> >> in the view. I really wouldn't mind not having to mention it anywhere
> >> inbetween.
>
> > This is actually a nice thing to do. Sorry that I misunderstood your
> > intentions :)
>
> >> Thank you for all your help,
> >> Clemens
>
> >> Clemens Oertel
> >> clem...@oertel.ca
>
> Clemens Oertel
> clem...@oertel.ca
>
> Clemens Oertel
> clem...@oertel.ca
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Lift" group.
To post to this group, send email to liftweb@googlegroups.com
To unsubscribe from this group, send email to
liftweb+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/liftweb?hl=en
-~----------~----~----~----~------~----~------~--~---
