From 266d94d2df845e4197e6e591e799c4c0fe6d9717 Mon Sep 17 00:00:00 2001
From: 0x80 <0x80@negativezero.link>
Date: Sat, 23 Apr 2022 00:00:00 +0000
Subject: [PATCH] Add support for funkwhale channels.

---
 src/determine_software.py   |  4 +++
 src/extractors/funkwhale.py | 50 +++++++++++++++++++++++++++++++++++++
 tests/funkwhale.txt         |  3 +++
 3 files changed, 57 insertions(+)
 create mode 100644 src/extractors/funkwhale.py
 create mode 100644 tests/funkwhale.txt

diff --git a/src/determine_software.py b/src/determine_software.py
index ce80a4a..e34d48b 100644
--- a/src/determine_software.py
+++ b/src/determine_software.py
@@ -7,6 +7,10 @@ supported_software = {
 		'<meta property="og:site_name" content="Bibliogram">',
 		'<section class="bibliogram-meta">'
 	],
+	"funkwhale" : [
+		'<meta name=generator content=Funkwhale>',
+		"<noscript><strong>We're sorry but Funkwhale doesn't work properly without JavaScript enabled. Please enable it to continue.</strong></noscript>"
+	],
 	"peertube" : [
 		'<meta property="og:platform" content="PeerTube"'
 	]
diff --git a/src/extractors/funkwhale.py b/src/extractors/funkwhale.py
new file mode 100644
index 0000000..48c28f5
--- /dev/null
+++ b/src/extractors/funkwhale.py
@@ -0,0 +1,50 @@
+#!/usr/bin/python3
+
+from utils import *
+
+# portable code to get filename
+import os
+platform = os.path.basename(__file__)
+if platform.endswith(".py"):
+	platform = platform[:(-3)]
+
+def extract_from_url (url, verbosity):
+	# split into domain and path
+	slash1 = url.find("/",url.find("//")+2)
+	domain = url[:slash1]
+	path = url[slash1:]
+
+	# get page type
+	slash2 = path.find("/",1)
+	if slash2 > 0:
+		page_type = path[1:slash2]
+	else:
+		page_type = path[1:]
+
+	if page_type == "channels":
+		slash3 = path.find("/",slash2 + 1)
+		if slash3 > 0:
+			channel_name = path[slash2+1:slash3]
+		else:
+			channel_name = path[slash2+1:]
+
+		# use home instance
+		if '@' in channel_name:
+			# TODO: how do we handle protocol (http vs. https)?
+			# for now, assume it's the same as url, or https if not specified
+			if "//" in domain:
+				protocol = domain[:domain.index("//")+2]
+			else:
+				protocol = "https://"
+			debug ("Assuming " + protocol + " for remote instance", verbosity, platform)
+			at_index = channel_name.find("@")
+			domain = protocol + channel_name[at_index+1:]
+			channel_name = channel_name[:at_index]
+
+		if channel_name != "":
+			return domain + "/api/v1/channels/" + channel_name + "/rss"
+
+def extract (url, page=None, network=False, verbosity=3, args={}):
+	feed = extract_from_url (url, verbosity)
+	if not feed is None:
+		return feed
diff --git a/tests/funkwhale.txt b/tests/funkwhale.txt
new file mode 100644
index 0000000..29f5ed9
--- /dev/null
+++ b/tests/funkwhale.txt
@@ -0,0 +1,3 @@
+https://funkwhale.it/channels/ilpunto/ https://funkwhale.it/api/v1/channels/ilpunto/rss
+https://funkwhale.thurk.org/channels/radio_colibri@open.audio/ https://open.audio/api/v1/channels/radio_colibri/rss
+https://open.audio/channels/radio_colibri https://open.audio/api/v1/channels/radio_colibri/rss