I took some time to look over the Architecture Review. Jaxer works much like ColdFusion (and other server side languages), where the Jaxer server will preprocess your code before returning it to the browser. Along with removing any server side code, Jaxer will add JavaScript code to your response to help enable it's basic feature sets. Read the review for more details.
How about what Jaxer supports on the server side? The docs say database access is supported (SQLite and MySQL), file access is supported, and network access. "Network Access" is a bit vague. It sounds like it could support POP or SMTP, but that the code would have to be written yourself.
The primary modification to your code seems to be the runat attribute added to script blocks. For example, this will log a message on the server:
<script runat="server">
Jaxer.Log.info("hi ray");
</script>
Another value for runat is server-proxy, which is rather cool. The meat of the functions will work on the server side, but the functions are callable from the front end. This is similar to the cfajaxproxy tag built in ColdFusion.
And lastly (again though - I'm guessing) is the "both" value. This lets you use a function both server side and client side. Here is an example from their sample code:
<!-- This script block will be available on both client and server since the
runat attribute is set to 'both'. -->
<script type="text/javascript" runat="both">
/*
* Easy access to a named element in the DOM
*/
function $(id)
{
return document.getElementById(id);
}
</script>
As I mentioned above, Jaxer will parse up your source code and return it modified. Using the example I mentioned earlier with the Logging, when I viewed the result in my browser, it was removed from the source.
