Acquired Opera AdBlock code

files/parse.js

Index: files/parse.js
===================================================================
new file mode 100644
--- /dev/null
+++ b/files/parse.js
@@ -0,0 +1,105 @@
+const parse = {/*
+	opera: function(content, add) {
+		//if(content.indexOf(';') == 0)//Comment
+		//	return;
+		if(add)
+			opera.extension.urlfilter.block.add(content);
+		else
+			opera.extension.urlfilter.block.remove(content);
+	},*/
+	adblock: function(content, add) {

[Felix Dahlke, 2012/10/05 15:47:16]

This function has a lot of code duplication, we should reduce that.

+		var url = this.wildcard(content.replace(/^@@/, '').replace(/\$.*$/, ''));
+		//opera.postError(content);
+		if(content.match(/\$/)) { //Is a type-specific rule
+			option = this.options(content.split('$')[1].split(','));
+			
+			if(content.match(/^@@/)) {
+				if(add)
+					opera.extension.urlfilter.allow.add(url, option);

[Felix Dahlke, 2012/10/05 15:47:16]

If we're using urlfilter.allow, filters with "adblock format" (e.g. Easylist)
will not work on Opera < 12.10. But even if we check for its presence to avoid
an exception, whitelisting wouldn't work, and that means we would have many
false positives.

+				else
+					opera.extension.urlfilter.allow.remove(url, option);
+			}
+			else {
+				if(add)
+					opera.extension.urlfilter.block.add(url, option);
+				else
+					opera.extension.urlfilter.block.remove(url, option);
+			}
+		}
+		else {
+			if(content.match(/^@@/)) {
+				if(add)
+					opera.extension.urlfilter.allow.add(url);

[Felix Dahlke, 2012/10/05 15:47:16]

See above.

+				else
+					opera.extension.urlfilter.allow.remove(url);
+			}
+			else {
+				if(add)
+					opera.extension.urlfilter.block.add(url);
+				else
+					opera.extension.urlfilter.block.remove(url);
+			}
+		}
+	},
+	map: {
+		'script': opera.extension.urlfilter.RESOURCE_SCRIPT,
+		'image': opera.extension.urlfilter.RESOURCE_IMAGE,
+		'stylesheet': opera.extension.urlfilter.RESOURCE_STYLESHEET,
+		'object': opera.extension.urlfilter.RESOURCE_OBJECT,
+		'xmlhttprequest': opera.extension.urlfilter.RESOURCE_XMLHTTPREQUEST,
+		'object-subrequest': opera.extension.urlfilter.RESOURCE_OBJECT_SUBREQUEST,
+		'subdocument': opera.extension.urlfilter.RESOURCE_SUBDOCUMENT,
+		'document': opera.extension.urlfilter.RESOURCE_DOCUMENT,
+		//'refresh': opera.extension.urlfilter.RESOURCE_REFRESH,
+		//'media': opera.extension.urlfilter.RESOURCE_MEDIA,
+		//'font': opera.extension.urlfilter.RESOURCE_FONT,
+		'other': opera.extension.urlfilter.RESOURCE_OTHER,
+		'all': opera.extension.urlfilter.RESOURCE_SCRIPT | opera.extension.urlfilter.RESOURCE_IMAGE | opera.extension.urlfilter.RESOURCE_STYLESHEET | opera.extension.urlfilter.RESOURCE_OBJECT |

[Felix Dahlke, 2012/10/05 15:47:16]

We should generate this 'all' mask from the flags above.

+			opera.extension.urlfilter.RESOURCE_XMLHTTPREQUEST | opera.extension.urlfilter.RESOURCE_OBJECT_SUBREQUEST | opera.extension.urlfilter.RESOURCE_SUBDOCUMENT | opera.extension.urlfilter.RESOURCE_DOCUMENT |
+			opera.extension.urlfilter.RESOURCE_REFRESH | opera.extension.urlfilter.RESOURCE_MEDIA | opera.extension.urlfilter.RESOURCE_FONT | opera.extension.urlfilter.RESOURCE_OTHER
+	},
+	options: function(special) {
+		var option = {excludeDomains: [], includeDomains: [], resources: this.map['all']};
+		var additive = false;
+		for(i in special) {
+			var element = special[i];
+			if(element.match(/^domain/i) == 0) { //TESTED

[Felix Dahlke, 2012/10/05 15:47:16]

What is "TESTED" supposed to mean? That this expression evaluates to true when
element starts with "domain"?  I don't think so, match() returns an array, which
isn't equal to 0, type inference or not. I don't think this condition is ever
entered.

+				var domains = special[i].split('=')[1].split(',');

[Felix Dahlke, 2012/10/05 15:47:16]

What if there is no "="? The above should match for /^domain=/ to avoid issues
with incorrect filters.

+				for(j in domains) {
+					if(domains[j].match(/^~/)) //Don't apply on this website

[Felix Dahlke, 2012/10/05 15:47:16]

This could be made more succinct by setting the option attribute conditionally.

+						option['excludeDomains'].push(domains[j]);
+					else //Apply on this website
+						option['includeDomains'].push(domains[j]);
+				}
+			}
+			else if(!element.match(/^elemhide/i)) { //No support for ElemHide - TESTED

[Felix Dahlke, 2012/10/05 15:47:16]

"TESTED" again? What about bogus or unsupported options? Whitelisting options
would be more robust.

+				if(!element.match(/^~/i)) {

[Felix Dahlke, 2012/10/05 15:47:16]

Case insensitive matching makes no sense for "~".

+					if(!additive) {
+						additive = true;
+						option['resources'] = 0;
+					}
+					option['resources'] += this.map[element];

[Felix Dahlke, 2012/10/05 15:47:16]

That's not how a bit flag is added to a mask ("|="). This won't work.

+				}
+				else if(!additive)
+					option['resources'] -= this.map[element];

[Felix Dahlke, 2012/10/05 15:47:16]

That's not how a bit flag is removed from a bit mask ("^="). This won't work.

+			}
+		}
+		return option;
+	},
+	wildcard: function(url) { //TESTED

[Felix Dahlke, 2012/10/05 15:47:16]

"TESTED" again?
There's a bit of duplication here and it's not easy to read, but the function
does seem to work fine.

+		//| at the beginning or end means no wildcard
+			
+		if(!url.match(/^\|\|/)) { //Does not starts with ||
+			if(url.match(/^\|/)) //Starts with |
+				url = url.replace(/^\|/, ''); //Remove the |
+			else if(!url.match(/^\*/)) //Avoid **
+				url = '*' + url; //Add the wildcard
+		}
+		if(url.match(/\|$/)) //Ends with |
+			url =  url.replace(/\|$/, ''); //Remove the |
+		else if(!url.match(/\*$/)) //Avoid **
+			url = url + '*'; //Add the wildcard
+		
+		return url;
+	}
+};