This is mostly just for my own notes, because this will come up later in the project.

To allow a plugin for spam modification in my plugin, I need to abstract out a few things:
The comment to be posted is given to an abstract checking function. This checking function in turn calls the plugin’s checking function. (This is like the sfGuard login function which can either run default or turn over login functionality to another function for such things as LDAP login).

The plugin has a few options for how it responds, all of which I should try to allow for:

1. It will set the comment as ham
2. It will set the comment as spam
3. It will run a secondary process to determine the humanity of the poster

After the comments are automatically dealt with, the user may change the results somewhat. The plugin will probably want to report on these changes.

1. User changed spam to ham (incorrectly marked as spam)
2. User changed ham to spam (incorrectly marked as ham)

There are probably three spam filters that I would make use of:

• Akismet
• Typepad AntiSpam
• Mollum

I should focus on these three and make certain that each one of them can be turned off/on with nothing more than changing the app’s config file.

This entry was posted on Saturday, February 6th, 2010 at 3:00 pm by NoCoolName_Tom and is filed under Uncategorized. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.