It would be helpful (without having to figure out how in the heck PCRE works) to have an option for having anything submitted in UTF-8 automatically flagged as probable spam.
I've got a large amount of Chinese language comment spam appearing in my approval queue and none of it is being caught by the spam module. In looking at the "edit comment" page for the comment (www.example.com/comment/edit/XXX), you can see that it is being submitted in UTF-8.
Since the average user (anonymous or registered) is likely to want to submit their comment in UTF-8, it seems to me that having a way to detect and assign a "probability" to comments submitted this way would be a valuable addition to this modules capabilities.
| Comment | File | Size | Author |
|---|---|---|---|
| UTF8commentspam.png | 18.32 KB | friskya |
Comments
Comment #1
friskya commentedWhoops!
I mean't "unlikely" to want to submit their comments in UTF-8
Sorry about that!
Comment #2
friskya commentedSorry - just discovered that my understanding of non-english language character display was wrong.
What I'm trying to say is, anything submitted in the format (& # 12362;) (had to add the whitespace to prevent it from displaying as it's Chinese equivalent) has a better than average chance of being spam, so an option to filter for anything submitted in that format would be helpful. I don't thing the average drupal user is going to know how to create a "perl compatible regular expression" right off.
Comment #3
jeremy commentedI am postponing this functionality until I have time to work on the 5.x-2.x version of the module.
Comment #4
jeremy commentedThe 5.x-3.x development branch allows for easier development of custom filters, such as this one. Re-opening against this development branch.
Comment #5
jeremy commentedPostponing this request until someone comes along with a patch. This should be implemented as a plug-in, not as a patch to the core spam module.
Comment #6
killes@www.drop.org commentedNobody was interested after all.