X-Git-Url: https://code.delx.au/youtube-cgi/blobdiff_plain/0353143a6d35c80ee24acd3cf8f3502743eaab56..7ee672be8998bfa85a8552f0e9c129d241e11f50:/youtube.cgi
diff --git a/youtube.cgi b/youtube.cgi
index b94febf..ef7b608 100755
--- a/youtube.cgi
+++ b/youtube.cgi
@@ -15,8 +15,7 @@ import urllib.parse
import urllib.request
-MAX_MEMORY_BYTES = 128 * 1024*1024
-USER_AGENT = "Mozilla/5.0 (X11; Ubuntu; Linux x86_64; rv:15.0) Gecko/20100101 Firefox/15.0.1"
+USER_AGENT = "Mozilla/5.0 (X11; Linux x86_64; rv:67.0) Gecko/20100101 Firefox/67.0"
MIMETYPES = {
"video/mp4": "mp4",
@@ -36,8 +35,11 @@ QUALITIES = {
class VideoUnavailable(Exception):
pass
+class NotYouTube(Exception):
+ pass
+
def print_form(url="", msg=""):
- script_url = "http://%s%s" % (os.environ["HTTP_HOST"], os.environ["REQUEST_URI"])
+ script_url = "https://%s%s" % (os.environ["HTTP_HOST"], os.environ["REQUEST_URI"])
sys.stdout.write("Content-Type: text/html\r\n\r\n")
sys.stdout.write("""
@@ -103,6 +105,16 @@ def urlopen(url, offset=None):
assert start == offset
return res
+def validate_url(url):
+ parsed_url = urllib.parse.urlparse(url)
+ scheme_ok = parsed_url.scheme == "https"
+ host_ok = parsed_url.netloc.lstrip("www.") in ["youtube.com", "youtu.be"]
+
+ if scheme_ok and host_ok:
+ return
+ else:
+ raise NotYouTube()
+
def parse_url(url, parser):
f = urlopen(url)
parser.feed(f.read().decode("utf-8"))
@@ -214,6 +226,7 @@ def decode_signature(js_url, signature):
p.stdin.close()
transformed_signature = p.stdout.read().decode("utf-8").strip()
+ transformed_signature = urllib.parse.unquote(transformed_signature)
if p.wait() != 0:
raise Exception("js failed to execute: %d" % p.returncode)
@@ -253,7 +266,8 @@ def get_best_video(player_config):
signature = None
if signature:
- video_url = append_to_qs(video_url, {"signature": signature})
+ sp = url_data.get("sp", ["signature"])[0]
+ video_url = append_to_qs(video_url, {sp: signature})
best_url = video_url
best_quality = quality
@@ -278,21 +292,24 @@ def get_video_url(page):
if not video_url:
return None, None
- filename = sanitize_filename(page.title)
- filename += "." + extension
+ title = player_config["args"].get("title", None)
+ if not title:
+ title = json.loads(player_config["args"]["player_response"])["videoDetails"]["title"]
+ if not title:
+ title = "Unknown title"
+
+ filename = sanitize_filename(title) + "." + extension
return video_url, filename
class YouTubeVideoPageParser(html.parser.HTMLParser):
def __init__(self):
super().__init__()
- self.title = None
self.unavailable_message = None
self.scripts = []
def handle_starttag(self, tag, attrs):
attrs = dict(attrs)
- self._handle_title(tag, attrs)
self._handle_unavailable_message(tag, attrs)
self._handle_script(tag, attrs)
@@ -302,13 +319,6 @@ class YouTubeVideoPageParser(html.parser.HTMLParser):
def _ignore_data(self, _):
pass
- def _handle_title(self, tag, attrs):
- if tag == "title":
- self.handle_data = self._handle_title_data
-
- def _handle_title_data(self, data):
- self.title = data.strip()
-
def _handle_unavailable_message(self, tag, attrs):
if attrs.get("id", None) == "unavailable-message":
self.handle_data = self._handle_unavailable_message_data
@@ -343,11 +353,12 @@ def cgimain():
try:
url = args["url"][0]
except:
- print_form(url="http://www.youtube.com/watch?v=FOOBAR")
+ print_form(url="https://www.youtube.com/watch?v=FOOBAR")
return
try:
page = YouTubeVideoPageParser()
+ validate_url(url)
parse_url(url, page)
video_url, filename = get_video_url(page)
video_data = urlopen(video_url)
@@ -356,10 +367,15 @@ def cgimain():
url=url,
msg="Sorry, there was an error: %s
" % cgi.escape(e.args[0])
)
+ except NotYouTube:
+ print_form(
+ url=url,
+ msg="Sorry, that does not look like a YouTube page!
"
+ )
except Exception as e:
print_form(
url=url,
- msg="Sorry, there was an error. Check your URL?
"
+ msg="Sorry, there was an unknown error.
"
)
return
@@ -411,7 +427,7 @@ def main():
try:
url = sys.argv[1]
except:
- print("Usage: %s http://youtube.com/watch?v=FOOBAR" % sys.argv[0], file=sys.stderr)
+ print("Usage: %s https://youtube.com/watch?v=FOOBAR" % sys.argv[0], file=sys.stderr)
sys.exit(1)
page = YouTubeVideoPageParser()