1
0
forked from jan6/bot6
bot6/youtube.py

172 lines
6.1 KiB
Python
Raw Permalink Normal View History

2021-10-05 13:18:25 +00:00
from html.parser import HTMLParser
from urllib.request import urlopen
class YouTube:
y, z = {}, {} # empty on every invokation
video_type = ""
2021-10-05 13:18:25 +00:00
def mesg(self, msg, t=None):
self.util.mesg(msg, t)
2022-08-23 14:11:47 +00:00
def match_urls(self, str):
r = [
i
for i in str.split()
if "https://youtu.be/" in i
or "https://www.youtube.com/watch?v=" in i
or "https://m.youtube.com/watch?v=" in i
or "https://youtube.com/watch?v=" in i
or "https://www.youtube.com/embed/" in i
or "https://www.youtube-nocookie.com/embed/" in i
2022-09-14 15:35:23 +00:00
or "https://music.youtube.com/watch?v=" in i
2022-09-26 21:16:34 +00:00
or "https://youtube.com/shorts/" in i
or "https://www.youtube.com/shorts/" in i
or "https://www.youtube.com/clip/" in i
or "https://youtube.com/clip/" in i
]
2022-08-23 14:11:47 +00:00
r = list(dict.fromkeys(r))
n = 0
for i in r:
if not i.startswith("http"):
r.pop(n)
n += 1
return r
def is_embed(str):
return str.startswith("https://www.youtube.com/embed/") or str.startswith(
"https://www.youtube-nocookie.com/embed/"
)
2022-09-14 15:35:23 +00:00
def is_ytmusic(str):
return str.startswith("https://music.youtube.com/watch?v=")
2022-09-26 21:16:34 +00:00
def is_ytshorts(str):
2022-09-26 22:11:51 +00:00
return str.startswith("https://youtube.com/shorts/") or str.startswith(
"https://www.youtube.com/shorts/"
)
2022-09-26 21:16:34 +00:00
def is_clip(str):
return str.startswith("https://youtube.com/clip/") or str.startswith(
"https://www.youtube.com/clip/"
)
2021-10-05 13:18:25 +00:00
class parseprop(HTMLParser):
def __init__(self):
print("yt parse init")
HTMLParser.__init__(self)
self.itemprops_list = ["name", "duration", "uploadDate", "interactionCount"]
if YouTube.video_type == "clip":
self.itemprops_list += ["description"]
print("it is a clip!")
2021-10-05 13:18:25 +00:00
def handle_starttag(self, tag, attrs):
if (tag != "meta" and tag != "link") or (
(
[i for i in attrs if "itemprop" in i] == []
and ("name", "title") not in attrs
)
or (tag == "meta" and ("itemprop", "name") in attrs)
):
2021-10-05 13:18:25 +00:00
return
global z
# print(self,tag,attrs)
2021-10-05 13:18:25 +00:00
for k, v in attrs:
if k == "itemprop":
if v not in self.itemprops_list:
2021-10-05 13:18:25 +00:00
return
x = [v]
if tag == "link" and v == "name":
x = ["channelName"]
2021-10-05 13:18:25 +00:00
elif k == "content":
if attrs[0][1] == "interactionCount":
v = int(v)
2021-10-05 13:18:25 +00:00
x += [v]
elif k == "name" and v == "title":
x = [v]
else:
return
2021-10-05 13:18:25 +00:00
z.update({x[0]: x[1]})
# print(x[0],"=",x[1])
def fmt_dur(dur):
h, m, s = 0, 0, 0
m = dur[2:].split("M")
s = int(m[1][:-1])
m = int(m[0])
if m >= 60:
h = m // 60
m = round((m / 60 - h) * 60)
return f"{h}h {m}m {s}s"
2022-02-07 16:28:19 +00:00
elif h == 0 and m == 0 and s == 0:
return "LIVE"
elif m == 0 and s != 0:
return f"{s}s"
elif s == 0:
return f"{m}m"
2021-10-05 13:18:25 +00:00
else:
return f"{m}m {s}s"
def yt(self, url):
# self.util.mesg("dbg hello")
2022-02-07 16:28:19 +00:00
url = url.rstrip("\x01")
self.video_type = (
"clip"
if self.is_clip(url)
else "shorts"
if self.is_ytshorts(url)
else "music"
if self.is_ytmusic(url)
else "embed"
if self.is_embed(url)
else "video"
)
video_type = self.video_type
if video_type == "embed":
videoId = url.split("/")[4]
url = f"https://www.youtube.com/watch?v={videoId}"
elif video_type == "music":
2022-09-14 15:35:23 +00:00
for i in url.split("?")[1].split("&"):
2022-09-26 22:11:51 +00:00
if i[0:2] == "v=":
videoId = i[2:]
2022-09-14 15:35:23 +00:00
url = f"https://www.youtube.com/watch?v={videoId}"
elif video_type == "shorts":
2022-09-26 22:11:51 +00:00
videoId = url.split("?")[0].split("/")[-1]
2022-09-26 21:16:34 +00:00
url = f"https://www.youtube.com/watch?v={videoId}"
2021-10-05 13:18:25 +00:00
global y, z
y, z = {}, {}
p = self.parseprop()
# use premature optimization? it should be SLIGHTLY faster
if self.premature_optimization:
url_h, data = urlopen(url), b""
# <body> appears on approximately line 21 or 22, so we read 24 lines to be safe (23-25 should be license comment)
# I tried to read byte amounts but it's hard to make sure no invalid utf8 bytes happen due to partial reads
for i in range(24):
data += url_h.readline()
data = data.decode() # bytes to utf-8
url_h.close()
else:
# just read all of the html
data = urlopen(url).read().decode()
# print(f"\x1b[31m my data is: {data}\x1b[0m")
2021-10-05 13:18:25 +00:00
p.feed(data)
if y == z == {}:
irc_string = "[\x0304Youtube\x03] \x0307ERROR:\x0308 got no data from server! \x0315(check your URL for typos!)\x03"
ansi_string = "[\x1b[31mYoutube\x1b[0m] \x1b[33;2mERROR:\x1b[33;1m got no data from server! \x1b[37;2m(check your URL for typos!)\x1b[0m"
print(ansi_string)
return irc_string, True
2021-10-05 13:18:25 +00:00
z.update({"duration": self.fmt_dur(z["duration"])})
y, z = z, {}
irc_string = f"[\x0303Youtube\x03] \x02{y['title']}\x02 ({y['duration']}) uploaded by \x1d{y['channelName']}\x1d on {y['uploadDate']}, {y['interactionCount']:,} views"
ansi_string = f"[\x1b[32mYoutube\x1b[0m] \x1b[1m{y['title']}\x1b[0m ({y['duration']}) uploaded by \x1b[03m{y['channelName']}\x1b[0m on {y['uploadDate']}, {y['interactionCount']:,} views"
2021-10-05 13:18:25 +00:00
print(ansi_string)
return irc_string, False
2022-08-23 14:11:47 +00:00
if __name__ == "__main__":
import sys
YouTube.premature_optimization = False
2022-08-23 14:11:47 +00:00
YouTube.yt(YouTube, sys.argv[1])