Very cool! This type of compression is something that I think should also available in Jesse's Java version. So the java version would have two compression levels to choose from.
Question: ------------ Any idea on how long it will take to decompress on the client? Suggestion: ------------ It would be cool if we can also create compression library for IOElement that will compress data sent from the server. In this way we could compress html, text, etc on-the-fly. -- Raymond Irving -- Raymond Irving --- Stephen Carroll <[EMAIL PROTECTED]> wrote: > I can agress to that, probably the socket connection > stays open, but the > thread from the HTTP server must still parse the > HTTP request and return the > data in a timeslice with everyone else, and that can > take time. I hope to > work on my compression schema soon enough here. > > I've taken a look at brainjar and found it > facinating and have also > attempted to correct for missing semi-colons, etc. > This process does appear > to be intensive and I'm not sure if I will include a > large amount of that > functionality. The compression schema I've worked on > simply goes through the > javascript source and breaks up content into > 'words', simply building a > dictionary of common terms, then analyzes how often > those phrases appear in > the code itself. Lastly, the calculation determins > if one would actually > save anything from using symbolic substitutions. For > example, the following > code: > > DynLayer.prototype.fadeIn = function(inc,ms){ > this._opacity=0; > this._fadeMode='in'; > this.fadeTo(100,(inc||4),ms); > }; > > would be reduced to: > > eval(Decompress("fade,this","DynLayer.prototype.~1In > = > ~F(inc,ms){~2._opacity=0;~2._~1Mode='in';~2.~1To(100,(inc||4),ms);};") > > You can see the 'dictionary' as the first parameter > in the Decompress > routine. The decompression routine would be served > up in only the first > request to dynapi.js, and perhaps a common > dictionary for most common > javascript keywords would be included like > 'function'. > > > ----- Original Message ----- > From: "Jesse Vitrone" <[EMAIL PROTECTED]> > To: "Stephen Carroll" <[EMAIL PROTECTED]> > Cc: <[EMAIL PROTECTED]> > Sent: Sunday, June 08, 2003 10:53 AM > Subject: Re: [Dynapi-Dev] Re: compressing files and > dynapi.asp (aka server > edition) > > > > Stephen, > > Like others have posted already, I think both > of our ideas can be > > good additions for DynAPI. > > I'm not sure about how HTTP keep alive works, > but when I have my > > page load in multiple scripts in Mozilla, you can > see the status bar > > showing Mozilla connecting / retrieving each file. > So having the page > > only load one large file sped up the page loading > a good deal. > > > > Jesse > > > > > > Stephen Carroll wrote: > > > > >LOL, well it seems like alot of us are on the > same thread of thought. I > like > > >the idea of not having to hit the server multiple > times, but doesn't > 'HTTP > > >keep alive' maintain one socket connection for > the duration of gifs, src, > > >etc. that are requested from a given HTTP server? > This removes the > overhead > > >of open/close sockets at the TCP/IP level to a > minimum for any given > client. > > >I do believe there is a speed increase in that > HTTP parsing doesn't have > to > > >take place if it can be included in one main > file. For library.load, I > > >introduced a delay to allow multiple files to be > requested from the HTTP > > >server in my dynapi.asp idea. I also like the > idea of never even having > to > > >mess with compression altogether, thats inherent > of the dynapi.asp > behavior. > > > > > > > > > > > > > > > >------------------------------------------------------- > > >This SF.net email is sponsored by: Etnus, makers > of TotalView, The best > > >thread debugger on the planet. Designed with > thread debugging features > > >you've never dreamed of, try TotalView 6 free at > www.etnus.com. > > >_______________________________________________ > > >Dynapi-Dev mailing list > > >[EMAIL PROTECTED] > > > >http://www.mail-archive.com/[EMAIL PROTECTED]/ > > > > > > > > > > > > > > > ------------------------------------------------------- > This SF.net email is sponsored by: Etnus, makers of > TotalView, The best > thread debugger on the planet. Designed with thread > debugging features > you've never dreamed of, try TotalView 6 free at > www.etnus.com. > _______________________________________________ > Dynapi-Dev mailing list > [EMAIL PROTECTED] > http://www.mail-archive.com/[EMAIL PROTECTED]/ __________________________________ Do you Yahoo!? Yahoo! Calendar - Free online calendar with sync to Outlook(TM). http://calendar.yahoo.com ------------------------------------------------------- This SF.net email is sponsored by: Etnus, makers of TotalView, The best thread debugger on the planet. Designed with thread debugging features you've never dreamed of, try TotalView 6 free at www.etnus.com. _______________________________________________ Dynapi-Dev mailing list [EMAIL PROTECTED] http://www.mail-archive.com/[EMAIL PROTECTED]/