#bouncer-filters

project / repo / mailing list / issues

A filter list generator for the Bouncer SMS filtering app.

ℹ️ Note
The canonical project locations are linked above. Other locations are mirrors.

#Usage

Get the latest output file from the repo.

Or, generate the output yourself:

1. Add words to the input list input_list.txt.

2. Run the program:

   python filter.py
   

3. Import the output file output/filter_list_*.json into Bouncer.

#Configuration

#Input

input_list.txt was created from a number of different 'common spam phrases' lists from across the internet. Modify it as you wish.

#Word lists

The word lists (word_lists/) are used to exclude common words from the block lists so that they aren't too sensitive. You can modify the existing word lists or add a new one.

word_lists/word_list_soft.txt is based on this word frequency list from Wikipedia and contains the top 2000 words from contemporary fiction.

word_lists/word_list_strict.txt is based on this word frequency list from first20hours/google-10000-english and contains a list of the 10,000 most common English words from Google's Trillion Word Corpus.

#Output

Two files are output, one for each word list described above. Choose whichever strictness you prefer.

#Contribution

Ideally, suggest changes using the Sourcehut mailing list. You can also create an issue on the Github Issues page.

#License

This repository is licensed under the MIT license. See LICENSE file.

The strict word list (word_lists/word_list_strict.txt) is licensed as follows:

Data files are derived from the Google Web Trillion Word Corpus, as 
described by Thorsten Brants and Alex Franz, and distributed by the 
Linguistic Data Consortium. Subsets of this corpus distributed by Peter 
Novig. Corpus editing and cleanup by Josh Kaufman.

Educational and personal/research use of this data is permitted under 
the LDC license, Norvig's MIT license for his contributions, and US 
fair use doctrine. I do not recommend using this data for commercial 
purposes without licensing it from the Linguistic Data Consortium.