From a560958c3adab652276b24e69f4ffd2794f1c7a3 Mon Sep 17 00:00:00 2001
From: Aevann <randomname42029@gmail.com>
Date: Mon, 23 Jan 2023 04:06:56 +0200
Subject: [PATCH] fix youtube timestamps

---
 files/helpers/regex.py    |  2 +-
 files/helpers/sanitize.py | 37 +++++++++++++++++++++++++------------
 files/routes/posts.py     | 17 ++---------------
 3 files changed, 28 insertions(+), 28 deletions(-)

diff --git a/files/helpers/regex.py b/files/helpers/regex.py
index 0eee98294..08a9b4282 100644
--- a/files/helpers/regex.py
+++ b/files/helpers/regex.py
@@ -85,7 +85,7 @@ imgur_regex = re.compile(f'(https:\/\/i\.imgur\.com\/[a-z0-9]+)\.({image_regex_e
 
 giphy_regex = re.compile('(https:\/\/media\.giphy\.com\/media\/[a-z0-9]+\/giphy)\.gif', flags=re.I|re.A)
 
-youtube_regex = re.compile('(<p>[^<]*)(https:\/\/youtube\.com\/watch\?v\=([a-z0-9-_]{5,20})[\w\-.#&/=\?@%+]*)', flags=re.I|re.A)
+youtube_regex = re.compile('(<p>[^<]*)(https:\/\/youtube\.com\/watch\?[\w\-.#&/=?@%+]{7,})', flags=re.I|re.A)
 yt_id_regex = re.compile('[a-z0-9-_]{5,20}', flags=re.I|re.A)
 
 link_fix_regex = re.compile("(\[.*?\]\()(?!http|\/)(.*?\))(?!([^<]*<\/(code|pre|a)>|[^`]*`))", flags=re.A)
diff --git a/files/helpers/sanitize.py b/files/helpers/sanitize.py
index d0176cc2f..7c98380b2 100644
--- a/files/helpers/sanitize.py
+++ b/files/helpers/sanitize.py
@@ -237,6 +237,27 @@ chud_images = listdir("files/assets/images/chud")
 chud_images = [f'![](/i/chud/{f})' for f in chud_images]
 chud_images.extend([':#trumpjaktalking:', ':#reposthorse:'])
 
+def handle_youtube(url):
+	html = None
+	params = parse_qs(urlparse(url).query, keep_blank_values=True)
+	id = params.get('v')[0]
+
+	t = None
+	split = id.split('?t=')
+	if len(split) == 2:
+		id = split[0]
+		t = split[1]
+
+	if yt_id_regex.fullmatch(id):
+		if not t:
+			t = params.get('t', params.get('start', [0]))[0]
+		if isinstance(t, str): t = t.replace('s','')
+		html = f'<lite-youtube videoid="{id}" params="autoplay=1&modestbranding=1'
+		if t:
+			html += f'&start={int(t)}'
+		html += '"></lite-youtube>'
+	return html
+
 @with_sigalrm_timeout(10)
 def sanitize(sanitized, golden=True, limit_pings=0, showmore=True, count_marseys=False, torture=False, sidebar=False, snappy=False):
 	sanitized = sanitized.strip()
@@ -356,17 +377,9 @@ def sanitize(sanitized, golden=True, limit_pings=0, showmore=True, count_marseys
 		if i.group(0) in captured: continue
 		captured.append(i.group(0))
 
-		params = parse_qs(urlparse(i.group(2)).query, keep_blank_values=True)
-		t = params.get('t', params.get('start', [0]))[0]
-		if isinstance(t, str): t = t.replace('s','')
-
-		htmlsource = f'{i.group(1)}<lite-youtube videoid="{i.group(3)}" params="autoplay=1&modestbranding=1'
-		if t:
-			try: htmlsource += f'&start={int(t)}'
-			except: pass
-		htmlsource += '"></lite-youtube>'
-
-		sanitized = sanitized.replace(i.group(0), htmlsource)
+		html = handle_youtube(i.group(0))
+		if html:
+			sanitized = sanitized.replace(i.group(0), html)
 
 	sanitized = video_sub_regex.sub(r'\1<p class="resizable"><video controls preload="none" src="\2"></video></p>', sanitized)
 	sanitized = audio_sub_regex.sub(r'\1<audio controls preload="none" src="\2"></audio>', sanitized)
@@ -472,12 +485,12 @@ def normalize_url(url):
 	url = url.replace("https://youtu.be/", "https://youtube.com/watch?v=") \
 			 .replace("https://music.youtube.com/watch?v=", "https://youtube.com/watch?v=") \
 			 .replace("https://www.youtube.com", "https://youtube.com") \
+			 .replace("https://m.youtube.com", "https://youtube.com") \
 			 .replace("https://youtube.com/shorts/", "https://youtube.com/watch?v=") \
 			 .replace("https://youtube.com/v/", "https://youtube.com/watch?v=") \
 			 .replace("https://mobile.twitter.com", "https://twitter.com") \
 			 .replace("https://m.facebook.com", "https://facebook.com") \
 			 .replace("https://m.wikipedia.org", "https://wikipedia.org") \
-			 .replace("https://m.youtube.com", "https://youtube.com") \
 			 .replace("https://www.twitter.com", "https://twitter.com") \
 			 .replace("https://www.instagram.com", "https://instagram.com") \
 			 .replace("https://www.tiktok.com", "https://tiktok.com") \
diff --git a/files/routes/posts.py b/files/routes/posts.py
index ff2769411..fe4eb9693 100644
--- a/files/routes/posts.py
+++ b/files/routes/posts.py
@@ -630,21 +630,8 @@ def submit_post(v:User, sub=None):
 				embed = requests.get("https://publish.twitter.com/oembed", params={"url":url, "omit_script":"t"}, timeout=5).json()["html"]
 				embed = embed.replace('<a href', '<a rel="nofollow noopener" href')
 			except: pass
-		elif url.startswith('https://youtube.com/watch?v='):
-			url = unquote(url).replace('?t', '&t')
-			yt_id = url.split('https://youtube.com/watch?v=')[1].split('&')[0].split('%')[0]
-
-			if yt_id_regex.fullmatch(yt_id):
-				params = parse_qs(urlparse(url).query, keep_blank_values=True)
-				t = params.get('t', params.get('start', [0]))[0]
-				if isinstance(t, str): t = t.replace('s','')
-
-				embed = f'<lite-youtube videoid="{yt_id}" params="autoplay=1&modestbranding=1'
-				if t:
-					try: embed += f'&start={int(t)}'
-					except: pass
-				embed += '"></lite-youtube>'
-
+		elif url.startswith('https://youtube.com/watch?'):
+			embed = handle_youtube(url)
 		elif SITE in domain and "/post/" in url and "context" not in url and url.count('/') < 6:
 			id = url.split("/post/")[1]
 			if "/" in id: id = id.split("/")[0]