What do you think of the efficiency of utlizing FSO (FileSystemObjects) to read in a textstream file of "matching criteria" for a filter? I envision being able to maintain a long set of REGEX matching criteria (A Blacklist) in a Windows text file. THe format of the file would be "one regex string per line" and I'd loop through the input using FSO's ReadLine method. This approach makes maintaining the list of filter criteria much easier than via the TBird3 filter manager interface. My concern is whether the resource and time efficiency of opening and parsing the file with Javascript for each incoming mail message will be too much overhead. I'd appreciate your input on this approach.
A similar capability would be possible if one could utilize an addressbook as a storage mechanism for this set regex's. Of course, I'd need to be able to specify that the "Is in Address Book" condition be looking at a field other than the email address (i.e. Notes or Customx fields).
My understanding is that the issues with js and filters are likely to be if you constantly switch between C++ objects and js objects, which you do not need to do. Javascript itself is considered now to be a first-class language for Mozilla development. Also, filtering of incoming messages is not really a time sensitive task (though it is on the main user interface thread, so you risk locking the interface if too slow.) But my gut feeling would be that opening a file, reading in a list of RegExes per message, and applying those to the subject would not be too slow. Applying that to the body is a different matter, though it is my understanding that SpamAssassin does that.
You might consider doing your own custom filter rather than relying on FiltaQuilla. The only thing hard is the XBL for the dropdown list.
Login 
Register 
Home
Print this Post 
Add a New Topic
Reply to Post
Quote and Reply 
Topic RSS 
