I am not sure if I understood your question. Let me try to answer anyway:
First of all you need to understand how the communication from user to user works. Assume we have only one single server with one channel and several users just to make it easier to understand. The principle for a huge network with thousands of users is the same:
1. The user sends a line from his IRC client to the channel and therefore to the server of that network.
2. The server checks the text and stores it in his buffer (usually 512 Bytes - the rest will be cut off).
3. The text is send back to the channel.
4. All users will see this text - some earlier and some later - including you.
All you can do is to
react on a event - but you can not prevent it from happen. This example also makes clear why such a think like a
Kick and/or Ban Protection is simply not possible.
Back to your example: All you can do is to
IGNORE the user -
AFTER the first "bad" line has apeared on your clients channel window.
You may be able to react on words or fragments of a sentences and take precautions - you just need a
trigger to make your event react upon.
Example for a swear word kick/ban and ignore:
ON @1:TEXT:*:#:{
IF (*badword* iswm $1-) {
ban $chan $nick 3
kick $chan $nick Increase Your Vocabulary!
ignore -u60 $nick
}
}
Hopefully this explains the background why your filter idea would not work.
