I wrote a shell script for automatically deleting CP spam when it's reported. Before I start using it, I want to check if anybody can find a way to make it misbehave.
It works like this: - Look through the report queue for a banned regex (typically the URL shortener they use) - If it's there, use the JSON interface to check the OPs of all reported threads for the regex - If a thread has the regex in its OP, delete it
What's to stop me from just reporting a thread I don't like as CP?
What's the plan for when they just use a different shortener? Add them to the banlist one by one?
It checks the text.
Fuck, I almost reported this thread
What are you planning on using for said regex?
How do I add this on Holla Forums?
For now, tr.im/. I can add other domains with \| as a separator using grep basic regex syntax.
If they stop using URLs in the post message (they did that with Lynxchan) I'll experiment with command line OCR tools. OCR got rid of them on Lynxhub and Endchan.
You run it with a cronjob on a server. If you can get me a volunteer account for Holla Forums I can run it there once it's tested.
I've been wanting to implement something like this to offer up for months and was too lazy to get to work. Thank you for this so much.
The problem with this is when they don't post any link shorteners in the text, only in the image. It happens a lot, if you recall correctly.
Also this solution is only working around the fact that the site is so fucking broken that you can't moderate it properly, and they apparently have no interest in fixing it, even going to far to implement an overboard out of the site's scope.
You wouldn't be the first one trying to crash this picture.
What a hot head. BLO BLO BLO BLO BLO BLO BLO BLO
I noticed /just/ often has the spam links broken. You might want to get in touch with its BO for some common link shorteners.
That's what OCR is for. Text recognition in images. For example: $ tesseract topbane.jpeg stdout Topbane . ruPSHC,PSHC-Big Guys Videos4U Yo - Mosqui‘ro Men