#!/usr/bin/env bash # Use rclone to sync the last two good Wikimedia XML data dumps. set -euxo pipefail if [ -z "$1" ]; then echo "Please supply a mirror URL" exit 1 fi MIRROR_URL="$1" RCLONE_REMOTE=$(basename "$MIRROR_URL") NTFY_IP=$(sudo tailscale status | grep "dietpi" | awk '{print $1}') wget "${MIRROR_URL}/rsync-filelist-last-2-good.txt" \ -O /home/jas/rsync-filelist-last-2-good.txt grep "enwiki" /home/jas/rsync-filelist-last-2-good.txt | grep -v "tenwiki" | tee /home/jas/rsync-filelist-last-2-good-en.txt rm -fv /home/jas/rsync-filelist-last-2-good.txt rclone sync \ --http-no-head \ --transfers 8 \ --include-from /home/jas/rsync-filelist-last-2-good-en.txt "${RCLONE_REMOTE}:" \ /naspool/archives/wikimedia-xmldatadumps-en rm -fv /home/jas/rsync-filelist-last-2-good-en.txt curl \ -H prio:default \ -H tags:incoming_envelope \ -d "Syncing of wikimedia xml datadumps succeeded" \ "http://${NTFY_IP}:8080/wikimedia_xmldatadumps_en" exit 0