THIS PROJECT IS DEAD.
- Script Started Sat Dec 26 00:20:17 UTC 2020
- Script Ended Sat Dec 26 00:34:39 UTC 2020
- Script Took 14 Minutes To Filter 132 Lists. See Log Here.
- The Edited AllParsed File is 42 MB And Contains 2077699 Domains.
- Average Parsing Of 37149.8 lines in 4 BlackLists was 25.75 Seconds.
- 1752 Valid Top Level Domains. No New TLD's.
- 6 Lists That Do NOT Use https
This Project Aims To Universally take ANY Blacklist, and ensure that it is formatted to be compatible with Pi-hole(tm)
- Lists update daily if there are any changes.
- Build a user-driven blacklist.
- Build a user-driven whitelist.
- Mirror and Filter, any user-submitted blacklist.
- Handle ANY list, even if it is compressed.
I've neglected this repository for a very long time. As of 26-12-2020, I've received notification fro github that the way this script updates won't be supported. I will leave this repo as-is for a period of time, but ultimately will remove it from github.
Due to it's unwieldy .git index size, I was forced to rm .git && git init
and force push the repo without any history. This gives me a clean slate to work with.
This script runs daily in a Proxmox LXC, and updates this repository automatically.
I may provide basic instructions on how to run this yourselves, but I cannot provide support for it. Honestly, with my daily cron, there is no need to run it yourself. The code is in the repo to be viewed, and is opensource. The provided lists can be downloaded securely from github.
Additionally, I do NOT condone the usage of the combined lists. Do so at your own discretion. It is literally EVERY list combined and deduped. If you like blocking your internet, or boasting a high blocking count, go for it.
- Individual lists tend to be safer than all of them Combined.
- You will find them Within the "Subscribable-Lists" directory.
- There are now Country Specific Lists!
Adding Them to Pi-hole(tm)
Simply copy the RAW format url for the list and add them.
- In the Web Interface on the Settings page.
- Note, I honestly don't recommend adding the big list, it WILL break websites.
Just add
https://raw.githubusercontent.com/deathbybandaid/piholeparser/master/Subscribable-Lists/CombinedBlacklists/CombinedBlackLists.txt
-
To request a list to be whitelisted or blacklisted, please submit an issue containing WHY it should be added or removed.
- I'm already parsing all of the lists daily and uploading them to the parsed directory in this repository.
- If you prefer to use this project yourself locally, Keep Reading.
Caution: The Script Has Evolved to the point that it runs other analytical tasks that add time to the process.
You have been warned.
There is a querylists.sh within the scripts directory.
This will allow me to query the individual parsed files for a specific domain.
There is a Log Available
This should provide some insights as to what lists are dead, empty, or too large for github.
A pun on Pi-hole's gravity.sh, this tool allows me to see what domains are on my list versus gravity.list
- Script updates first thing on every run, always the most up to date version.
- Script Checks for dependencies.
- .lst files are named on purpose to help name the end results better.
- Script skips steps if the file is empty
- Script skips IP Lists (for now)
- Script appends RecentRunLog to tell me that a list is no longer dead.
- Script Pushes the results to localhost, and Github (if selected).
- Script runs daily with cron, or Manually.
- Allparsed list is based on the userbase.
- Checks to see if host of list is available.
- Checks to see if a list was updated online.
- Download based on host availability, file extension (tar or 7z), or attempt to use a mirrored copy from this repository.
- Creates a mirror if file is not empty, or over the Github 100MB limit.
- Remove Commented lines #'s !'s and Empty Lines.
- Remove Invalid Characters. FQDN's are allowed to use dashes, underscores, and emoji's. all other symbols are not allowed.
- Remove Pipes | and Carots ^
- Removes IP Addresses.
- Remove Empty Space.
- Checks for FQDN Requirements. A Period and a Letter.
- Remove Periods at Beginning and End Of Lines.
- Filter out common file extensions used in assets
- Reverse Searches Top Level Domains
- Remove Duplicates, If any.
- Create Parsed File, if it survives this process.
- This will take all the small lists and merge them.
- I then take that list, add user-submitted blacklists, remove user-submitted whitelists, and produce another Big List.
- I take the Big List and generate small lists based on Country Codes.
All "Original Unaltered Lists" are located within the mirroredlists directory.
After going through the parser, many lists contain zero lines and are deleted.
The filtered lists are in the parsed directory;
with filenames to reflect the Original Creators work/effort.