Topic: account/login/LoginAction.do

Ignore me, cookies were disabled in the browser ... sorted now. :)

I'm curious, how are you managing to use javascript to login?

I mucked around a while back logging into the api using javascript on a private webpage. Due to browser security (cross domain checking) I could only do it by using adding the greasemonkey extension to firefox and using its GM_xmlhttpRequest function.

fred,

ahh yes, cross domain checking ... funny how that one caught me out ...

I've been teaching myself web app technology recently (mainly been java 'fat client' or server process developer in the past) and didn't read about the cross domain security at first.
And to begin with I was testing my html/javascript by loading it directly into the browser from the local file system (ie file:///Users/blah rather than html://localhostblah) ... which seems to allow you to then use XMLHttpRequest to call any domain you like.

So of course I wrote a whole bunch of stuff directly calling the various betfair .do scripts. Lovely.

Then I took the next step so I could get a bit more sophisticated and write my own servlets. I installed glassfish and bundled my web app and deployed it.
Then when I tested the same stuff from http:/localhost:8080/myWebApp it stopped working!!!!!

Took me a while to figure out what was going on and was pretty p1ssed off when I realised, thinking I'd wasted all my time!

However as a workaround what I've done is write a few servlets of my own which act as a proxy to betfair. So i can call https://localhost:8080/myWebApp/account/loginAction.do for example.

Not as big a pain as it sounds as the proxy servlet is very simple, but it obviously adds an extra couple of network hops into the equation so won't be using it for critical apps!

Must admit I'm quite enjoying dabbling with webapp development. You can get some quite nifty stuff running quickly using AJAX techniques.

Ok, I'm in a generous mood so I'll drop a couple of useful links in:

http://www.prototypejs.org
http://https://glassfish.dev.java.net

brichy,

You don't want to be using java.net.URL, you want to be using HttpClient.
See http://hc.apache.org/httpclient-3.x

The tutorial on there should get you going. (http://hc.apache.org/httpclient-3.x/tutorial.html)
It's really simple and cookies are auto handled by default I think.

As for decompressing gziped stuff, take a look at class java.util.zip.GZIPInputStream

Thanks Mike. I was just curious, I'm not actively pursuing a javascript solution now, it was just something I was looking at the other year as an unusual bot programming language that anybody could use (everybody has a web browser and text editor) without having to download/configure/workaround anything.