redarc

A self-hosted solution to search, view and archive link aggregators.

Supports:

• Reddit
• HackerNews (in progress)

Features:

• Ingest pushshift dumps
• View threads/comments
• Fulltext search via PostgresFTS
• Submit threads to be archived via API
• Periodically fetch rising, new and hot threads from specified subreddits
• Download i.redd.it images from threads.

Please abide by the Reddit Terms of Service and User Agreement if you are using their API

Download pushshift dumps

https://the-eye.eu/redarcs/

All data 2005-06 to 2022-12:

magnet:?xt=urn:btih:7c0645c94321311bb05bd879ddee4d0eba08aaee&tr=https%3A%2F%2Facademictorrents.com%2Fannounce.php&tr=udp%3A%2F%2Ftracker.coppersurfer.tk%3A6969&tr=udp%3A%2F%2Ftracker.opentrackr.org%3A1337%2Fannounce

Top 20,000 subreddits:

magnet:?xt=urn:btih:c398a571976c78d346c325bd75c47b82edf6124e&tr=https%3A%2F%2Facademictorrents.com%2Fannounce.php&tr=udp%3A%2F%2Ftracker.coppersurfer.tk%3A6969&tr=udp%3A%2F%2Ftracker.opentrackr.org%3A1337%2Fannounce

Installation:

Master branch is unstable. Please checkout a release

Docker

Install Docker: https://docs.docker.com/engine/install

Services:

• postgres: Main database for threads, comments and subreddits
• postgres_fts: Database for full-text searching
• redarc: API backend and React frontend
  • Requires: redis, reddit_worker if INGEST_ENABLED
• redis: Required for any service that uses a task queue
• image_downloader: Asynchronously downloads images from Reddit if DOWNLOAD_IMAGES
  • Requires: redis, reddit_worker
• index_worker: Indexes threads/comments into postgres_fts
  • Requires: postgres_fts and postgres
• reddit_worker: Asynchronously fetches threads/comments from Reddit
  • Requires: redis, image_downloader
• subreddit_worker: Asynchronously fetches hot/new/rising thread IDs from subreddits
  • Requires: reddit_worker and redis

If you wish to change the postgres password, make sure POSTGRES_PASSWORD and PGPASSWORD are the same.

If you are using redarc on your personal machine, set docker envars REDARC_API=http://localhost/api and SERVER_NAME=localhost.

REDARC_API is the URL of your API server; it must end with /api eg: http://redarc.mysite.org/api.

REDARC_FE_API is the URL of the API server you want the frontend to send requests to.
If you are not using a reverse proxy, it should be the same as REDARC_API.

SERVER_NAME is the URL your redarc instance is running on. eg: redarc.mysite.org

Setting an INGEST_PASSWORD and ADMIN_PASSWORD in your API is highly recommended to prevent abuse.

IMAGE_PATH is the path you want image_downloader worker to download images. This is the same path the API backend fetches images from.

INDEX_DELAY is how often you want index_worker to index comments/threads

SUBREDDITS is a list of subreddits you want subreddit_worker to fetch threads from. It is delimited by commas

FETCH_DELAY is how often you subreddit_worker to fetch threads.

NUM_THREADS is the number of threads you want downloaded from hot, rising or new.

Docker compose (Recommended):

Docker compose:

Modify envars as needed

$ git clone https://github.com/Yakabuff/redarc.git
$ cd redarc
$ git fetch --all --tags
$ git checkout tags/vx.y.z -b vx.y.z
// Modify .env as-needed
$ cp default.env .env
$ docker compose up -d

Manual installation:

$ git clone https://github.com/Yakabuff/redarc.git
$ cd redarc

1) Provision Postgres database

$ docker pull postgres
$ docker run \
  --name pgsql-dev \
  -e POSTGRES_PASSWORD=test1234 \
  -d \
  -v postgres-docker:/var/lib/postgresql/data \
  -p 5432:5432 postgres 

$ docker run \
  --name pgsql-fts \
  -e POSTGRES_PASSWORD=test1234 \
  -d \
  -v postgresfts-docker:/var/lib/postgresql/data \
  -p 5433:5432 postgres 

psql -h localhost -U postgres -a -f scripts/db_submissions.sql
psql -h localhost -U postgres -a -f scripts/db_comments.sql
psql -h localhost -U postgres -a -f scripts/db_subreddits.sql
psql -h localhost -U postgres -a -f scripts/db_submissions_index.sql
psql -h localhost -U postgres -a -f scripts/db_comments_index.sql
psql -h localhost -U postgres -a -f scripts/db_status_comments.sql
psql -h localhost -U postgres -a -f scripts/db_status_comments_index.sql
psql -h localhost -U postgres -a -f scripts/db_status_submissions.sql
psql -h localhost -U postgres -a -f scripts/db_status_submissions_index.sql
psql -h localhost -U postgres -p 5433 -a -f scripts/db_fts.sql
psql -h localhost -U postgres -a -f scripts/db_progress.sql

2) Process dump and insert rows into postgres database with the load_sub/load_comments scripts

Note: Be sure the ingest and Reddit workers are disabled

python3 scripts/load_sub.py <path_to_submission_file>
python3 scripts/load_comments.py <path_to_comment_file>
python3 scripts/load_sub_fts.py <path_to_submission_file>
python3 scripts/load_comments_fts.py <path_to_comment_file>
python3 scripts/index.py [subreddit_name]
python3 scripts/unlist.py <subreddit> <true|false>

3) Start the API server.

$ cd api
$ python -m venv venv
$ source venv/bin/activate
$ pip install gunicorn
$ pip install falcon
$ pip install rq
$ pip install python-dotenv
$ pip install psycopg2-binary
$ gunicorn app

4) Start the frontend

cd ../redarc-frontend
mv sample.env .env

Set address for API server in the .env file

VITE_API_DOMAIN=http://my-api-server.com/api/

npm i
npm run dev // Dev server

5) Provision NGINX (Optional)

Edit nginx/nginx_original.conf with your own values

$ cd ..
$ mv nginx/redarc_original.conf /etc/nginx/conf.d/redarc.conf

cd redarc-frontend
npm run build 
cp -R dist/* /var/www/html/redarc/
systemctl restart nginx

6) Setup submission workers

Fill in .env files with your own credentials.

$ docker pull redis
$ docker run --name some-redis -d redis
$ cd redarc/ingest
$ python -m venv venv
$ source venv/bin/activate
$ pip install rq
$ pip install python-dotenv
$ pip install praw
$ pip install psycopg2-binary
$ pip install gallery-dl
$ python3 ingest/reddit_worker/reddit_worker.py
$ python3 ingest/index_worker/index_worker.py
$ python3 ingest/subreddit_worker/subreddit_worker.py
$ python3 ingest/image_downloader/image_downloader.py

Ingest data:

Postgres:

Note: Be sure the ingest and Reddit workers are disabled

Ensure python3, pip and psycopg2-binary are installed:

# Decompress dumps

$ unzstd <submission_file>.zst

$ unzstd <comment_file>.zst

$ pip install pyscopg2-binary

# Change database credentials if needed

$ python3 scripts/load_sub.py <path_to_submission_file>

$ python3 scripts/load_sub_fts.py <path_to_submission_file>

$ python3 scripts/load_comments.py <path_to_comment_file>

$ python3 scripts/load_comments_fts.py <path_to_comment_file>

$ python3 scripts/index.py [subreddit_name]

# Optional
$ python3 scripts/unlist.py <subreddit> <true|false>
$ python3 scripts/backfill_images.py <subreddit> <after timestamp utc> <num urls>

Web:

• Submit Reddit URL using the web form /submit to be fetched by reddit_worker
• Add subreddits to the SUBREDDITS envar (delimited by commas) to be periodically fetched by subreddit_worker

API:

search/comments?

• [unflatten = <True/False>]
• [subreddit = <name>]
• [id = <id>]
• [before = <utc_timestamp>]
• [after = <utc_timestamp>]
• [parent_id = <parent_id>]
• [link_id = <link_id>]
• [sort = <ASC/DESC>]

search/submissions?

• [subreddit = <name>]
• [id = <id>]
• [before = <utc_timestamp>]
• [after = <utc_timestamp>]
• [sort = <ASC|DESC>]

search/subreddits

search?

• <subreddit = <subreddit>>
• [before = <unix timestamp>]
• [after = <unix timestamp>]
• [sort = <asc|desc>]
• [query = <seach phrase>]
• <type = <comment|submission>>

License:

Redarc is licensed under the MIT license