Caching is used to speed up a call to display a template file by saving its contents and/or dynamic parts to a cache service. If a cached version of the call is available, that is returned instead of reading it directly from extra storage.
Hanjst server-side components are able to cache their raw contents of template files as so to accelerate the speed of generating outputs for HTTP requests.
Regarding to GWA2 in Java, caching service is provided with its core components of WebApp.class, here follows an example. (eg08131113)
//- suppose Hanjst extends WebApp, and be ready
Hanjst hanjst = new Hanjst();
String rawTemplate = hanjst.readTemplate(templateFile);
//- inside Hanjst.readTemplate to enable cache
//- suppose Wht is ready
public String readTemplate(String mytpl){
String tplcont = "";
HashMap hmArgs = Wht.initHashMap("file,enablecache",
mytpl+",1");
HashMap hmtmp = this.getBy("file:", null, hmArgs);
if((boolean)hmtmp.get(0)){
tplcont = (String)hmtmp.get(1);
tplcont = this.replacePath(tplcont, viewdir);
}
return tplcont;
}
In the example above, the enablecache tells GWA2 core components to try its cache service firstly for the targeted object. This kind of services are also available for database service, session service, etc.
However, since templates are dynamic, it is important to be careful what we are caching and for how long. For instance, if we are displaying the front page of a website that does not change its content very frequently, it might work well to cache this page for a few hours.
On the other hand, if we are displaying a page with a timetable containing new information by the minute, it would not make sense to cache this page. Cache is time-sensitive and some contents are good, but others not.
@todo