distroscraper 1.0.2

Scrape Linux Distro torrents and add to Transmission automatically

distroscraper

Scrapes Linux distribution web pages for new torrent files and adds them to Transmission. I want to give back to the Free Software community and one way I can do that is by seeding their torrents.

For legitimate redistrbution of permitted content only.

Included are scrapers for these distributions:

• Arch Linux
• Debian
• Fedora
• Linux Mint
• Raspberry Pi

Would love to include, if they would provide a legitimate torrent:

• OPNsense

If your favorite distribution is not mentioned, you can open an issue, or contribute a pull request. Take a look at the other files in /distroscraper/scrape and make it look like those in general.

Installation

Create and configure a Python Virtual Environment for distroscraper:

python -m venv /home/myser/.venv_distroscraper
. /home/myuser/.venv_distroscraper/bin/activate
pip install distroscraper

Create a shell script (I call mine "scrape-and-add.sh") to run the scrapers and pipe output through the transmission torrent adder. For the below script you will also need the 'jq' command line JSON processing tool.

#!/bin/bash

VENV_PATH=/path/to/venv
DISTROS="distroscraper.scrape.archlinux
         distroscraper.scrape.centos
         distroscraper.scrape.debian
         distroscraper.scrape.fedora
         distroscraper.scrape.linuxmint
         distroscraper.scrape.raspberrypi"
TM_ARGS="--host localhost
         --port 9091
         --user myuser
         --password supersecret
         --download-dir /path/to/linuxdownloads"

for d in $DISTROS; do
  $VENV_PATH/bin/python -m $d | \
    jq .href | \
    xargs $VENV_PATH/bin/python -m distroscraper.tm_add_torrent $TM_ARGS
done

Create a systemd timer or cron job to run scrape-and-add.sh periodically. Transmission is smart enough not to add the same torrent multiple times.

Here is an example systemd service file for use with a timer. The file name is intended to be /etc/systemd/system/distroscraper.service.

[Unit]
Description=scrape linux distributions

[Service]
User=service
Group=service
Type=oneshot
Nice=19
ExecStart=/home/service/scrape-and-add.sh

And here is the corresponding timer file. Note the randomized delay ensures that these happen "more or less" daily, but that everyone isn't hitting the download pages simultaneously at midnight. Call this file /etc/systemd/system/distroscraper.timer.

[Unit]
Description=scrape distributions daily

[Timer]
OnCalendar=00:00
RandomizedDelaySec=86400
Persistent=true
Unit=distroscraper.service

[Install]
WantedBy=timers.target

Reporting

If you want to ingest transmissions stats into some logging solution like ELK or Graylog, you can use this which will produce some JSON that can be massaged into these systems:

python -m distroscraper.tm_get_stats

For Graylog, you may be interested in my [GELF JSON Flattener|https://jefftickle.com/projects/gelf-flattener]