On Aug 25, 9:26 pm, David Pollak <[email protected]>
wrote:
> On Tue, Aug 25, 2009 at 11:09 AM, marius d. <[email protected]> wrote:
>
> > David,
>
> > The action attribute is no different than class, id and target
> > attributes currently supported when lift is magically building the
> > form tag around the snippet content. My other notes inline.
>
> > Br's,
> > Marius
>
> > On Aug 25, 8:11 pm, David Pollak <[email protected]>
> > wrote:
> > > On Tue, Aug 25, 2009 at 9:46 AM, marius d. <[email protected]>
> > wrote:
>
> > > > The topic was brought up in the discussion.
>
> > > > Br's,
> > > > Marius
>
> > > > On Aug 25, 7:32 pm, David Pollak <[email protected]>
> > > > wrote:
> > > > > On Tue, Aug 25, 2009 at 9:23 AM, marius d. <[email protected]>
> > > > wrote:
>
> > > > > > 1. I just pushed a little change that allows:
>
> > > > > > <lift:MySnippet form="post" action="/search">
> > > > > > ...
> > > > > > </lift:MySnippet>
>
> > > > > > 2. You can also do
>
> > > > > > <form method="post" action="/search">
> > > > > > <lift:mysnippet>
>
> > > > > > <f:inputSearch/>
> > > > > > <f:submit/>
>
> > > > > > </mysnippet>
>
> > > > > > </form>
>
> > > > > I'm not overly keen on the change (I'm not asking for it to be
> > removed).
> > > > > The reason for the "form" attribute was that it's common to do
> > postbacks
> > > > and
> > > > > the generation of S.uri as part of the URL was a bunch of
> > boilerplate.
>
> > > > This commit does not contradict this rule. People can use action only
> > > > if they really need to without the boiler plate of typing the form
> > > > tag.
>
> > > Right, but it's a place where Lift is mutating the result of the snippet
> > > call. This should only be done if there is a compelling reason. I see
> > no
> > > compelling reason here. This case is a fringe case and can easily be
> > > handled with a static form tag. The reason for the form attribute is
> > that
> > > it's the common case for wrapping a snippet in a post-back form and the
> > URL
> > > of the postback would require yet another snippet to calculate... and the
> > > feature made it into Lift before attribute snippets.
>
> > > So, the rules are: (1) Lift does not mutate what the snippet returns
> > except
> > > under extremely compelling situations and (2) we don't add "reserved
> > word"
> > > attributes except under compelling situations.
>
> > > > > That's the reason I made the exception to the "snippets result in
> > > > whatever
> > > > > is returned from the snippet rather than having magic intervention"
> > rule.
>
> > > > I really don't see how this implies a "magic intervention" since all
> > > > this applies only if user explicitly adds action attribute. I mean if
> > > > users explicitly points to a new page they expect that page to be
> > > > rendered naturally which is what is happening.
>
> > > Except for the "form='xxx'" attribute, I know of no other attribute
> > that's
> > > baked into Lift's snippet handling mechanism. The use of this attribute
> > is
> > > the only time that I know of that Lift's snippet handler mutates the
> > result
> > > of the snippet by putting a <form/> tag around it.
>
> > > In all other cases, it's up to the snippet itself to interpret the
> > > attributes. If we keep adding attributes that cause something to happen
> > if
> > > the attribute appears, it will reduce the attributes available for use by
> > > snippets themselves.
>
> > > > > In the future, let's not mutate what the snippet returns without a
> > > > > discussion and a very strong justification.
>
> > > > I really don't see the problem. The form attribute tells the user that
> > > > a form will be rendered. Since a form will be rendered why can't the
> > > > user change the action in exceptional cases since this doesn't induce
> > > > side effects. If it does please let me know what are those side
> > > > effects.
>
> > > It removes an attribute that's available for generic use by snippets. If
> > > there's code out there that uses the action attribute as part of the
> > > snippet, what are the impacts on that snippet?
>
> > There is no impact really. Current behavior is not altered in any way.
> > If user uses the action attribute he/she explicitly tells "I just what
> > to got to this URI instead of the default one". If user doesn't use
> > action attribut it'll render S.uri as it currently does. If there are
> > form snippets out there using the action attribute has no impact of
> > the snippet per se but we'll put that value to the action of the form
> > which indeed will not be ok. Still I really doubt that there are
> > snippets using form attribute that uses action attribute for a
> > different meaning that for a normal html form. Same question can apply
> > for current supported attributed class, id and target and that's ok. I
> > haven't seen people complaining that they use the class attribut
> > expressing something else then a CSS class. I think people are very
> > comfortable and knowledgeable in terms of HTML semantics and they just
> > don't mix these semantics with something else.
>
> I was unaware of these attributes being co-opted by the form handler. Under
> the standard:
>
> (1) Lift does not mutate what the snippet returns except under extremely
> compelling situations and (2) we don't add "reserved word" attributes except
> under compelling situations.
>
> The copying of id and class (and probably style) would meet the test. It's
> extremely compelling because almost every form element will have a class
> and/or style attribute. Many will have an id attribute.
>
> The number of times one has a target of a post-back is limited and I would
> have objected to the use of the "target" attribute in this case.
target attribute was needed for a concrete case. Ajax like file
upload. Because we can't upload files via Ajax the trick is to use a
hidden iframe and set the target for the form. This has been quite
helpful
>
> I still object to this change (although I am not asking you to back it
> out.) Altering the action attribute of the form is not compelling. It's a
> fringe use-case with a very simple work-around.
>
> I am setting down the explicit rule that any other changes to the way we
> handle this special case and any time that we reserve an attribute (whether
> or not it makes sense from an HTML perspective), I expect that there will be
> a lot more discussion and analysis.
I feel that this is a much more subjective discussion then an
objective one. As you have such strong feelings about the matter I
will back it out as I care more about Lift then some egocentric
motivation and I really don't feel comfortable fighting for such minor
things.
>
> David
>
>
>
>
>
> > > > If you want to I have no problem removing it but I really don't see
> > > > what the issue is.
>
> > > > > Thanks,
>
> > > > > David
>
> > > > > > In both cases you can use Lift's html abstractions such as
> > SHtml.text
> > > > > > etc. and bind your function. In the second example we used a
> > snippet
> > > > > > that has no idea about form ... it is just wrapped by a form. So in
> > > > > > both cases you don't really need to call S.attr to manually get
> > > > > > parameters.
>
> > > > > > Br's,
> > > > > > Marius
>
> > > > > > On Aug 25, 6:24 pm, David Pollak <[email protected]>
> > > > > > wrote:
> > > > > > > On Tue, Aug 25, 2009 at 12:38 AM, Ulises <
> > [email protected]>
> > > > > > wrote:
>
> > > > > > > > Hey people,
>
> > > > > > > > I'm trying to have a form be submitted to a different page (so,
> > not
> > > > a
> > > > > > > > POST/GET back). The scenario is a search form which renders on
> > a
> > > > front
> > > > > > > > page as well as on the search page.
>
> > > > > > > > E.g.:
>
> > > > > > > > /index
> > > > > > > > should include a form that submits to /search
>
> > > > > > > > /search
> > > > > > > > form that submits to itself
>
> > > > > > > <form method="post" action="/search"><input type="text"
> > name="search"
> > > > > > > value=""/><input type="submit" value="Search"/></form>
>
> > > > > > > In the snippet that handles the search page:
>
> > > > > > > val searchTerm = S.param("search") openOr {S.error("no search
> > term");
> > > > > > > S.redirectTo(S.referer openOr "/")}
>
> > > > > > > > Any hints on how to do this/were to look to learn how to do
> > this? I
> > > > > > > > guess that the form on the front page could be just static HTML
> > > > > > > > however I'm not entirely sure how /search would deal with the
> > form
> > > > > > > > submission from a static form ...
>
> > > > > > > > Cheers,
>
> > > > > > > > Ulises
>
> > > > > > > --
> > > > > > > Lift, the simply functional web frameworkhttp://liftweb.net
> > > > > > > Beginning Scalahttp://www.apress.com/book/view/1430219890
> > > > > > > Follow me:http://twitter.com/dpp
> > > > > > > Git some:http://github.com/dpp
>
> > > > > --
> > > > > Lift, the simply functional web frameworkhttp://liftweb.net
> > > > > Beginning Scalahttp://www.apress.com/book/view/1430219890
> > > > > Follow me:http://twitter.com/dpp
> > > > > Git some:http://github.com/dpp
>
> > > --
> > > Lift, the simply functional web frameworkhttp://liftweb.net
> > > Beginning Scalahttp://www.apress.com/book/view/1430219890
> > > Follow me:http://twitter.com/dpp
> > > Git some:http://github.com/dpp
>
> --
> Lift, the simply functional web frameworkhttp://liftweb.net
> Beginning Scalahttp://www.apress.com/book/view/1430219890
> Follow me:http://twitter.com/dpp
> Git some:http://github.com/dpp
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"Lift" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/liftweb?hl=en
-~----------~----~----~----~------~----~------~--~---