From c0ae248e62704a9bd465a656a48b9a3aab903099 Mon Sep 17 00:00:00 2001
From: Alexander Bocken <alexander@bocken.org>
Date: Sat, 26 Jun 2021 20:35:43 +0200
Subject: [PATCH] added peertube support

---
 README.md |  9 +++++++--
 ripper    | 15 +++++++++++++++
 2 files changed, 22 insertions(+), 2 deletions(-)

diff --git a/README.md b/README.md
index b3a4dab..f130938 100644
--- a/README.md
+++ b/README.md
@@ -1,15 +1,21 @@
 # ripper
-Checks YT Channels and Bitchute channels for new videos and downloads them quickly
+Checks YouTube, Bitchute, and Peertube channels for new videos and downloads them quickly.
 
 This is considerably faster than giving `youtube-dl` the channel URIs directly.
+## YouTube
 This script can use the YouTube Data API v3 if available, but has a fallback if either:
 
 - no API Key is given
 - the Data quota for the day has been exceeded
 
+## Bitchute
 Bitchute can also be used with this script.
 This part of the script is quite generic and could easily be adjusted for other websites.
 
+## Peertube
+Peertube instances require the channel RSS feed URL to work.
+
+
 See the documentation inside the `ripper` file for more information.
 
 ## Install
@@ -22,4 +28,3 @@ After `git clone`ing this repo, run the helper `installer`
 ## Features to come
 
 - Odysee support
-- Peertube support -> could just be done via RSS feeds
diff --git a/ripper b/ripper
index 3c39553..dfd68f8 100755
--- a/ripper
+++ b/ripper
@@ -105,6 +105,20 @@ for channel in $BC_CHANNELS; do
 done
 grep 'bitchute' "$DLARCHIVE" | sed 's/bitchute /https:\/\/www\.bitchute\.com\/video\//' >> /tmp/alreadydownloaded$$
 
+##PEERTUBE
+echo "Scanning Peertube instances..."
+PT_CHANNELS="$(grep -Ev '(bichute|youtube)' $CHANNELSFILE | grep -v '^#')"
+for channel in $PT_CHANNELS; do
+	printf 'PT channel: %s\n' "$channel"
+	channelxml="$(curl -s "$channel")"
+	todownload="$(echo "$channelxml" | xml2 | grep '^/rss/channel/item/link=' | sed 's|^/rss/channel/item/link=||' | sort -u | grep -vf "$BLACKLIST" | tee -ai /tmp/todownload$$)"
+	#might not be always correct for all peertube instances?
+	channelinstance="$(echo "$channelxml" | xml2 | grep '^/rss/channel/link=' | sed 's|^/rss/channel/link=||')"
+	#creates fake video links for other instances but shouldn't be an issue
+	grep -E 'peertube' "$DLARCHIVE" | sed "s|peertube |$channelinstance/videos/watch/|" | grep "$todownload" >> /tmp/alreadydownloaded$$
+done
+
+
 ##DOWNLOAD VIDEOS FROM ACCUMULATED LINKS
 grep -vf /tmp/alreadydownloaded$$ /tmp/todownload$$ | sort -u > /tmp/new_videos$$
 rm -f /tmp/alreadydownloaded$$ /tmp/todownload$$
@@ -121,6 +135,7 @@ if [ "$number" -gt 0 ]; then
 	notify-send "Channel Ripper" "Finished downloading"
 fi
 
+
 if [ "$number" -eq 0 ]; then
 	echo "No new videos"
 fi