Taking Data from The INTERNET

hello

smart man i SL made the BABLER that use the google translation page for SL transltion chat.

the page of google is: http://www.google.com/translate_t

now, how he did it? how can you search in the internet and get data from there?
in the google page you need to write the sentence and press button to show the translate... how can you do that???

thanks in advance!

The mysteries of the internet.

It is actually quite simple.

The whole thing is a form and said form is submitted to a particular web address when you press that button. So instead of pressing that button you directly call the appropriate website/webpage with the required data.

And thou shalt be rewarded with a more or less (flawless) translation.

I'm guessing they have just worked out the protocol / interface used by the page and are posting a request to the google translator (or an equivelent) and receiving it back it the same manner that the original page would.

EDIT : Squirrel beat me to it.

It's actually more like a series of tubes.

well... what commands i can use? would you please direct me to WIKI page or something? i want to learn how to do it....

many thank

llHTTPRequest

One of the problems you'll run into, is that llHTTPRequest will only return a limited number of characters from the http return (2048 I think). And these days, there is so much stuff in an http page or form, that the result you're trying to parse down to tends to be beyond that limit.

So things like Babler tend to also use an intermediary server, maybe running a PHP script, which can receive results > 2048, parses it down to the final answer, and then returns it to you.

right...

something like this would work, need to have CURL

$url = "http://URL HERE";
$data = "data=" . trim($_GET['text']);  // the post form
//$data = htmlspecialchars($data);
$data = str_replace ( ' ', '%20', $data );
$data = str_replace(array("&","'"),array("%26","%27"),$data);

$ch = curl_init();   
curl_setopt($ch, CURLOPT_URL,$url); 
curl_setopt($ch, CURLOPT_POST, 1);
curl_setopt($ch, CURLOPT_POSTFIELDS, $data);
curl_setopt($ch, CURLOPT_RETURNTRANSFER, 1);
$output_data = curl_exec ($ch);    
curl_close ($ch);

that will get you the returned page. You'll need to know the form variables and how to craft what your trying to translate into the form post. Then you need to get the data you really need

function get_meanlingful($head_seperator,$buffer,$foot_seperator){
// GET THE MEANLINGFUL PART
        set_time_limit(120);
        list($head,$meanlingful)=split("$head_seperator",$buffer,2);
        list($meanlingful,$foot)=split("$foot_seperator",$meanlingful,2);
        
        return($meanlingful);
}

$head_separator="<TEXT BEFORE THE DATA YOU NEED>";
$foot_separator="<TEXT AFTER THE DATA YOU NEED";
$truedata = get_meanlingful($head_separator,$output_data,$foot_separator);

i use the above code for a translating product, so it would work for what you need