On Sun, Dec 6, 2009 at 12:37 AM, Koen Deforche <[email protected]> wrote: > Hey Richard, > > 2009/12/2 Richard Dale <[email protected]>: >> One of the reasons I haven't been working on Wt::Ruby is because I >> been doing some JavaScript bindings for Qt based on the same 'Smoke >> libs' that Wt::Ruby uses. >> >> http://gitorious.org/qtscript-smoke/qtscript-smoke >> >> Because of that I don't think it would be too hard to adapt the >> Wt::Ruby code to work with Qt QtScript, which is going to be based on >> the WebKit JavaScrpt engine in Qt 4.6. Is that a crazy idea? Does it >> make sense to have a service side framework based on JavaScript? I >> don't know. The Chrome V8 based Node.js server side framework has been >> generating a lot of excitement, and maybe a QtCore/Wt JavaScript stack >> would be an interesting combination.. > > It's certainly mind-bending :-) > > I certainly believe it would make sense. It has the potential for > defining your event handling in JavaScript which may be reworded into > a client-side JavaScript method (and thus run entirely in client-side > JavaScript). > > Imagine that you write something like: > > onClick = function (w, event) { > var i; > for (i = 0; i < w.parent().count(); ++i) { // WContainerWidget::count() > if (w.widget(i) == w) // WContainerWidget::widget() > w.hide(); > else > w.show(); > } > }; > > Now if you do this in Wt as a C++ method you will run this on the > server and have an update such as: hide("o14"); show("o17"); > > But you could indicate that you want to have this as a client-side > method, in which case you could rewrite it also in terms of DOM API: > > onClick = function (o, event) { > var i; > for (i = 0; i < o.parentNode.childNodes.length; ++i) { > if (o.childNodes[i] == o) > o.style.display = "none"; > else > o.style.display = "block"; > } > }; > > That would be a more powerful variant of the current automatic > JavaScript learning which only works for "stateless methods", applied > to arbitrary methods. > > I only wonder with these server-side JavaScript frameworks is how as a > developer you do not get confused about where runs what code and not > make the mistake of trying to call a server-side function from > client-side code and vice versa. This problem is similar to using GWT > where you need to keep the client-side Java and server-side Java well > separated.
Also, I am curious, why the webkit javascript and not tracemonkey or v8 as they tend to be a *lot* more tested and both are a lot faster (v8 with javascript that is designed more like classes, and tracemonkey for everything else)? ------------------------------------------------------------------------------ Join us December 9, 2009 for the Red Hat Virtual Experience, a free event focused on virtualization and cloud computing. Attend in-depth sessions from your desk. Your couch. Anywhere. http://p.sf.net/sfu/redhat-sfdev2dev _______________________________________________ witty-interest mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/witty-interest
