@John: Will do - I'll post a plug-in later this week.
On Mar 2, 6:30 am, John Resig <jere...@gmail.com> wrote:
> Kevin -
>
> Something like this we'd like to test out in the realm of plugins
> first - just release your work as a plugin (be sure to toss it up on
> plugins.jquery.com and link to some demos) and if people really start
> to use it we'll definitely consider it for core. That's generally how
> we evaluate most code that goes in to core (look at how plugins
> handled the problems first, then refine them).
>
> --John
>
>
>
> On Sun, Mar 1, 2009 at 5:19 PM, Kevin Dalman <kevin.dal...@gmail.com> wrote:
>
> > I am working on a project and need URL param-parsing, as I usually do.
> > But this time I decided to try something new - based on ColdFusion
> > syntax that I have used for ages.
>
> > ColdFusion creates a system-level hash structure named "URL" that
> > contains all the URL params. This is very convenient, so I created a
> > jQuery extension to do the same thing. It creates and populates a
> > "$.url" hash on-load. This is a static var similar to $.browser.
>
> > The $.url object does the *opposite* of the $.serialize method:
> > $.serialize converts complex data TO an URL format, while $.url
> > creates a data object FROM the URL params.
>
> > My initial version was very small - only a few bytes of code is needed
> > to parse simple parameters. This would address the basic needs of most
> > users. So I think this *at a minimum* would be a valuable addition to
> > the jQuery core. The code is very small, creates no conflicts, and
> > takes barely 1ms.
>
> > After I created the basic method, I added more elaborate parsing to
> > store numbers and booleans in their proper format, and to allow
> > 'complex data' to be passed - ie, arrays and hashes:
>
> > page.html?actors=[Eastwood,Bronson,Heston]
> > page.html?actor={first: Clint, last: Eastwood}
>
> > This includes automatic array creation when a param key is repeated:
>
> > page.html?actor=Eastwood&actor=Bronson&actor=Heston
>
> > ...becomes: actor=[ Eastwood, Bronson, Heston ]
>
> > This also allows arrays-of-arrays and arrays-of-hashes:
>
> > page.html?actor={first: Clint, last: Eastwood}&actor={first: Charles,
> > last: Bronson}
>
> > ...becomes:
>
> > actor = [
> > 0: {
> > first: 'Clint'
> > , last: 'Eastwood'
> > }
> > 1: {
> > first: 'Charles'
> > , last: 'Bronson'
> > }
> > ]
>
> > You can see and test a demo page here...
>
> > http://layout.jquery-dev.net/url_parsing.html
>
> > There are a number of test URLs (hyperlinks) provided to demonstrate
> > the different types of parsing, but you can append any params you want
> > to the URL to see how they are parsed.
>
> > To illustrate the size of the 'long version', here is the partially
> > minified code:
>
> > // MINIFIED CODE (860 bytes)
> > function setURL(){
> > $.url={};$.urlParams=[];
> > var s=self.location.search.substr(1),p,d,k,v,i;
> > if(!s)return;
> > p=s.split("&");
> > for(i=0;i<p.length;i++){
> > d=p[i].split("=");k=$.trim(d[0]);
> > if(k){
> > v=d[1]==undefined?true:parse(d[1]);
> > if(!$.url[k]){$.url[k]=v;$.urlParams.push(k);}
> > else{if(!$.isArray($.url[k])||($.isArray(v)&&typeof $.url[k][0]!
> > ='object'))$.url[k]=[$.url[k]];$.url[k].push(v);}
> > }
> > }
> > function parse(x){
> > x=$.trim(x);
> > if(!x)return "";
> > var c=x.length-1,f=x.charAt(0),l=x.charAt
> > (c),A=f=="["&&l=="]",H=f=="{"&&l=="}",d,h,k,o,i;
> > if(A||H){
> > o=A?[]:{};d=x.substr(1,c-1).split(",")
> > for(i=0;i<d.length;i++){
> > if(A)o[i]=parse(d[i]);
> > else if(d[i]){h=d[i].split(":");k=$.trim(h[0]);if(k)o[k]=parse(h[1]);}
> > }
> > return o;
> > }
> > else if(!isNaN(x))return Number(x);
> > else if(x==="true")return true;
> > else if(x==="false")return false;
> > else return x;
> > }
> > }
>
> > The demo page contains a more readable, commented version of the code
> > above. If there is any interst in this code, feel free to help
> > yourself. I did not keep a copy of the short-version, but it would not
> > take long to recreate - this is not complex code.
>
> > SO, do John and the gang feel this addition would be worthwhile for
> > jQuery? I'm suggesting this partly out of self-interest - I copy the
> > same URL-parsing functions to every project I work on. I'd prefer that
> > this basic functionality was part of jQuery, and I feel a $.url object
> > is the most intuitive and flexible way to do it.
>
> > At a minimum, this will become part of my standard jQuery extensions
> > library. I prefer working with an 'URL object' rather than using a
> > 'parsing method':
>
> > // using an URL Object
> > if ($.url.section) doSomething( $.url.section );
>
> > // using a Parsing Method
> > var section = parseURL('section');
> > if (section) doSomething( section );
>
> > The ability to use complex objects offers more options for passing JS
> > data between pages:
>
> > $.each($.url.actor, function (idx, Actor) {
> > $('#List').append('<li>'+ Actor.first +' '+ Actor.last +'</li>');
> > });
>
> > The example above is a little silly, but you see how it could be
> > useful for passing 'state' or other data.
>
> > Feedback?
>
> > /Kevin- Hide quoted text -
>
> - Show quoted text -
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"jQuery Development" group.
To post to this group, send email to jquery-dev@googlegroups.com
To unsubscribe from this group, send email to
jquery-dev+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/jquery-dev?hl=en
-~----------~----~----~----~------~----~------~--~---
