> How big is the page?  At some point soon we may have the optimization issue
> worked out, but in reality the optimization wont substantially change the
> parse time, maybe shave off a second for an 'average' sized page.   The
> catch 22 is that the html 5 parser is the reason we can't optimize yet and
> if you are really crawling html in the wild you wouldn't get very far
> without it.  Rhino is slower than its C counterparts, which is why envjs
> developers have a number of other supported platforms at various stages of
> readiness.

> In general the platform you choose will be based on the project
> requirements,  if you dont need access to java apis, you might want to ask
> steven parkes or nickg about their integrations with spider monkey.

> For env.rhino.js your best solution in the near term is to simply thread a
> number of crawlers/scrappers to optimize processing time.

> Thatcher

> On Wed, May 5, 2010 at 3:26 PM, Cefn <google....@cefn.com> wrote:
> > Im using the rhino/js.jar distributed with env.js, and I found some
> > bizarre behaviour from invocation with any optimization turned on.

> > If I run the following script, js/bt/test.js

> > load("js/envjs/dist/env.rhino.js");
> > load("js/libs/sizzle.js");
> > Envjs("http://alistapart.com/", {
> >    logLevel: Envjs.DEBUG,
> >    scriptTypes: {
> >       'text/javascript': true
> >    }
> > });
> > (function(){
> >        var results = Sizzle("h4.title a");
> >        for(count = 0; count < results.length; count++){
> >                print(results[count].childNodes[0].data);
> >        }
> > })();

> > using the harness athttp://pastebin.com/zAmJpKNC(note the line
> > setOptimizationLevel(-1);) it correctly reports the titles of
> > Alistapart. A typical invocation is...

> > java -cp bin:js.jar Scraper

> > ...where bin contains the Scraper class, js.jar is rhino, and js/envjs/
> > dist/ contains the contents of the dist/ directory from envjs.

> > If I comment out the setOptimization(-1); line, then it throws errors
> > about missing javascript references, and doesn't even get to load the
> > page. Here's an example error it throws, just from changing that line.

> > java -cp bin:js.jar Scraper
> > Exception in thread "main" org.mozilla.javascript.EcmaError:
> > ReferenceError: "Envjs" is not defined. (test.js#2)

> > This is obviously fatal to using the framework with any acceleration.
> > The problem with running it unoptimized as a scraper, is that it takes
> > 2.5 seconds to complete the parse operation alone (barring env.js and
> > sizzle.js startup costs).

> > It would mean a lot to have it run optimized, which is why I tried
> > Sizzle instead of JQuery - to see if I could come in under the 64Kb
> > limit and turn optimization back on.

> > Perhaps I can avoid the env.js and sizzle.js startup costs for page+1,
> > page+2 at different URLs by just reusing the same Rhino Context, and
> > experts on this list may know a good way to do this, I'd welcome
> > suggestions, expecially given I may have to run this framework using
> > its slowest possible setting.

> > Anyone know what my options are to get JQuery-style selectors in
> > env.js without taking more than two seconds to process a single page.
> > I have hundreds of pages to process.

> > --
> > You received this message because you are subscribed to the Google Groups
> > "Env.js" group.
> > To post to this group, send email to envjs@googlegroups.com.
> > To unsubscribe from this group, send email to
> > envjs+unsubscribe@googlegroups.com <envjs%2Bunsubscribe@googlegroups.com>.
> > For more options, visit this group at
> >http://groups.google.com/group/envjs?hl=en.

> --
> Christopher Thatcher

> --
> You received this message because you are subscribed to the Google Groups "Env.js" group.
> To post to this group, send email to envjs@googlegroups.com.
> To unsubscribe from this group, send email to envjs+unsubscribe@googlegroups.com.
> For more options, visit this group athttp://groups.google.com/group/envjs?hl=en.