Yeah sure - thats the kind of functionality im after. Perhaps not quite so involved with java in xml, but sure, the variable screens and validations are what i'm trying to achieve for sure. Am I crazy / re-inventing the wheel to want to do something like this with lift?
Wow, awesome. SiteMap is easily the dark horse of lift :-)
> I was one of the folks who designed and built SmartMode back in
> 2000/2001. SiteMap in Lift is substantially what we built in SmartMode
> (but in SmartMode, we needed XML files rather than a DSL.)
When you say "SiteMap in Lift is substantially what we built in
SmartMode", are you talking conceptually, or functionally? If
functionally, then wow, seriously wow. Very cool. I'll dig into the
lift source to take a look at SiteMap tonight.....
I would be seriously keen to work on this. If you can provide a
> I've got a to-do hanging out to do Wizard as a Scala DSL in Lift... but
> I've got some documentation to write first. :-)
starting point or tips on where / how this might work I would happily
put all my "lift time" into making this work. Im also doing some cool
lift coding at work now, and this is something we'd really like so
could possibly commit some work days too.
Cheers
Tim
Interesting - can you elaborate on lifts flexibility over SmartMode?
> Both conceptually and functionally. SiteMap in SmartMode was more brittle
> than SiteMap in Lift. There are a few missing features from SmartMode
> SiteMap in Lift's implementation, but for the most part, it's stuff that
> wasn't used for the SmartMode powered sites.
It would be interesting to hear the comparison
Ok I will do :)
> I am enclosing the XML file that defined the Wizard XML. Maybe you can
> start modeling the DSL grammar based on the tags in Wizard.
Would I be correct in assuming that SiteMap gives you the flexibility
of display what and when conditionally, but as each form is being
generated dynamically, how would you commit back after each step - or
is it a case that all that data is held in state somehow and then it
just spits it out at the end to which you do what ever need be with
it? And if thats the case, how would it do that? Many objects, or one
massive lump of data etc?
Also, the stuff thats going to be broken away from mapper,
specifically the validation stuff - would you see that as being
reusable in this context?
Sorry for the barrage of questions - I just find this fascinating!
Cheers
Tim
Until they invent virtual beer, how's about IM?
> Each page in Wizard will have a call-back function and on success
> (at the very end of the Wizard when one does the final 'okay'), all
> the functions will be called and working with Mapper, there'll be a
> single Database transaction for all the function calls.
Given all the attention JPA is getting, that says to me that we would
need a solution that was persistence agnostic as much as possible.
This will obviously make it a lot tricker, but I think its really
important as the people who generally want this functionality, are
more "enterprisy" and are usually using stuff like Hibernate or
TopLink etc
> Yes.
Thought as much.
Cheers
Tim
> Over a beer, yes. :-)
Until they invent virtual beer, how's about IM?
Given all the attention JPA is getting, that says to me that we would
> Each page in Wizard will have a call-back function and on success
> (at the very end of the Wizard when one does the final 'okay'), all
> the functions will be called and working with Mapper, there'll be a
> single Database transaction for all the function calls.
need a solution that was persistence agnostic as much as possible.
This will obviously make it a lot tricker, but I think its really
important as the people who generally want this functionality, are
more "enterprisy" and are usually using stuff like Hibernate or
TopLink etc
> Yes.
Thought as much.
Cheers
Tim