Scratch

bobbybee wrote:

Can you open-source this? It looks interesting.

Sure! I'll do it tomorrow when I'm on my other computer.
Meanwhile, if it helps, I mainly used the REFERER superglobal to check out the page the user was on, and then simply looked for traces of unparsed BBCode    It's relatively short code.

bobbybee wrote:

I see...
Wait, how would it detect anti-dote if they're are no images?

It doesn't detect antidote, it detects whether the page needs antidote or not. To detect the actual antidote I would have to use JS too (which I cannot, from a PHP image).

That is pretty neat! I would love to have a look at the code for it and I think this is a pretty neat usage of dynamic signatures! I might add it to my API thread as a link at some point 

How does antidote change page HTML? Perhaps antidote could be detected by checking for [img ] tags? If there are no [img ] tags it would mean either there are no images, or antidote is installed and has replaced all the tags with <img> tags?

sparks,

The problem/aspect I came across during the testing is that the image, being a PHP one, cannot check the contents of the page the user is on. That would mean a big security hole for any browser! Therefore I had to get the page server-side only to check if the original contained BBCode -- not whether it had been parsed.

Javascript is essential to check the user's side, and unfortunately JS cannot be injected through images. I had been thinking about linking to a page that contained img tags, so that the user could check whether their browser effectively supported Antidote, but of course antidote only parses the code if it's on the scratch forums.

I hope that makes sense, because it doesn't to me 

Haha, antidote has converted most of that code to images    I'd better quote it to read.

How about making an antidote update so that antidote also checks for img tags on a page on your server? That might then be readable.

That is surprisingly simple to code! It takes brains to work out it can be done though, well done!

Last edited by sparks (2012-03-23 07:58:26)

Hardmath123 wrote:

?

I'm using Antidote but the image still pops up...

Yes, as I explained it the post to sparks, the image can only check whether there is BBCode, not whether the code has been parsed or not!

Ah, you see I didn't know that there was an antidote update! I shall go download it 

What you need is a way to detect the version    Does the firefox addon system allow for AJAX? You could use that to send the version number to a php file that checks to see if a newer version is avaliable and display a message with an update link?

By the way, if you're more of a PHPer like me, you might like this little function I built for Javascript to simulate PHP's file_get_contents():

function file_get_contents(url){
    if(window.XMLHttpRequest){
        request = new XMLHttpRequest();
    }
    else if(window.ActiveXObject){
        request = new ActiveXObject("Microsoft.XMLHTTP");
    }
    request.open("GET", url, false);
    request.send(null);
    return request.responseText;
}

It just lets me use AJAX in a method I'm more familiar with 

EDIT: Updated, that's better 

Last edited by sparks (2012-03-23 08:38:39)

cool make it say

img uploading...

anidite finding imgs

sparks wrote:

Ah, you see I didn't know that there was an antidote update! I shall go download it 

What you need is a way to detect the version    Does the firefox addon system allow for AJAX? You could use that to send the version number to a php file that checks to see if a newer version is avaliable and display a message with an update link?

By the way, if you're more of a PHPer like me, you might like this little function I built for Javascript to simulate PHP's file_get_contents():

Code:

function file_get_contents(url){
    if(window.XMLHttpRequest){
        request = new XMLHttpRequest();
    }
    else if(window.ActiveXObject){
        request = new ActiveXObject("Microsoft.XMLHTTP");
    }
    request.open("GET", url, false);
    request.send(null);
    return request.responseText;
}

It just lets me use AJAX in a method I'm more familiar with 

EDIT: Updated, that's better 

Wouldn't it be better to do:

function file_get_contents(url){
    if(window.XMLHttpRequest){
        request = new XMLHttpRequest();
    }
    else if(window.ActiveXObject){
        request = new ActiveXObject("Microsoft.XMLHTTP");
    }
    else {
        return false;
    }
    request.open("GET", url, false);
    request.send(null);
    return request.responseText;
}

That way if neither method exists (I have no idea when that would be true, but hey    ) it still works and doesn't error - or at least it errors nicely.

I suppose that through my use of compiled languages as opposed to interpreted languages, I have learned that error handling is of great importance    - otherwise it just crashes.

Last edited by rookwood101 (2012-03-23 18:04:21)

rookwood101 wrote:

sparks wrote:

Ah, you see I didn't know that there was an antidote update! I shall go download it 

What you need is a way to detect the version    Does the firefox addon system allow for AJAX? You could use that to send the version number to a php file that checks to see if a newer version is avaliable and display a message with an update link?

By the way, if you're more of a PHPer like me, you might like this little function I built for Javascript to simulate PHP's file_get_contents():

Code:

function file_get_contents(url){
    if(window.XMLHttpRequest){
        request = new XMLHttpRequest();
    }
    else if(window.ActiveXObject){
        request = new ActiveXObject("Microsoft.XMLHTTP");
    }
    request.open("GET", url, false);
    request.send(null);
    return request.responseText;
}

It just lets me use AJAX in a method I'm more familiar with 

EDIT: Updated, that's better 

Wouldn't it be better to do:

Code:

function file_get_contents(url){
    if(window.XMLHttpRequest){
        request = new XMLHttpRequest();
    }
    else if(window.ActiveXObject){
        request = new ActiveXObject("Microsoft.XMLHTTP");
    }
    else {
        return false;
    }
    request.open("GET", url, false);
    request.send(null);
    return request.responseText;
}

That way if neither method exists (I have no idea when that would be true, but hey    ) it still works and doesn't error - or at least it errors nicely.

I suppose that through my use of compiled languages as opposed to interpreted languages, I have learned that error handling is of great importance    - otherwise it just crashes.

Wait, hold on: this only works with local files, right? (I mean files on the same server) I thought this was a JS safety feature... 

Last edited by Hardmath123 (2012-03-24 03:55:18)

Hardmath123 wrote:

Wait, hold on: this only works with local files, right? (I mean files on the same server) I thought this was a JS safety feature... 

That's right. Unless the other server has expressedly allowed connection from the local server, in which case a connection can be made.

Squawkers13 wrote:

the image dissapeared

indeed.