RegexMatchError: __init__: could not find match for ^\w+\W #1918
Comments
|
Thank you for contributing to PyTube. Please remember to reference Contributing.md |
|
Same issue here, worked earlier today, and had this error for the past few hours over different machines. Seems youtube may have changed something to break PyTube. |
|
same issue here! Works fine till yesterday (from my side): venv/lib/python3.10/site-packages/pytube/cipher.py", line 33, in init |
|
same issue here! It seems like there's an issue with pytube. I've been using this library to work on my graduation project, and I would greatly appreciate it if you could fix it as soon as possible. It's really urgent. |
|
+1 on this issue |
|
+1 today |
|
my graduation project discussion is in 4 days please fix it as soon as possible |
|
+1 on this |
|
+1, let's thank youtube for changing their function names 🙏 |
|
Same error here: My logs in docker. |
|
just as a suggestion, my playlist is only like 2+hours so i downloaded all of it and bypass pytube for now. Work have to be done and i cant wait any longer. |
|
I very quickly migrated to https://github.com/yt-dlp/yt-dlp which seems more mature anyway. I am using it like this (using a named temporary directory and then returning bytes to be rendered by my request, adjust it to your own needs): from yt_dlp import YoutubeDL, DownloadError
def download_youtube_mp3(youtube_id, temporary_directory):
url = f"https://www.youtube.com/watch?v={youtube_id}"
ydl_opts = {
'format': 'bestaudio/best',
"max_filesize": 20 * 1024 * 1024,
"outtmpl": f"{temporary_directory}/%(id)s.%(ext)s",
"noplaylist": True,
"verbose": True,
'postprocessors': [{
'key': 'FFmpegExtractAudio',
'preferredcodec': 'mp3',
'preferredquality': '192',
}],
}
ydl = YoutubeDL(ydl_opts)
try:
meta = ydl.extract_info(
url,
download=True,
)
except DownloadError as e:
raise e
else:
video_id = meta["id"]
return {
"title": meta["title"],
"file_path": f"{temporary_directory}/{video_id}.mp3"
} with tempfile.TemporaryDirectory() as temporary_directory:
audio_details = download_youtube_mp3(youtube_id, temporary_directory)
with open(audio_details["file_path"], "rb") as f:
bytes_data = f.read()
return bytes_data |
|
+1 |
|
Found a quick fix - turns out current function name includes "$O", a non-word character which will not match "\w+". Edited to include @ghrist8p's comment: as per ECMA's spec, valid identifiers may include symbols "$" and "_" in addition to word-characters. Change cipher.py line 30 to: I do not know all the possible patterns, but this is working for me. |
Worked. |
how would I get it to work in Colab? |
|
As per the spec an identifier can have a |
Thank you for your help! I'm also looking forward to official updates from the program side. |
bless you |
|
Where did you enter var_regex = re.compile(r"^[\w$_]+\W")? It wasn't in the pytube code I was using. I'm getting an error in this code:
|
cipher.py on line 30. You can check my repo commit. |
Thank you for your help. |
|
i fix this bug today. |
|
+1 |
|
I've tried all of the above codes, but they don't work. They worked until yesterday. Working on Colab. |
@coco1718 var_regex = re.compile(r"^[\w$_]+\W") still works for me somehow |
|
I think there's another problem, but I can't find it |
|
I solved with this code: import requests
def download_stream(url, filename):
"""
Simulates downloading a video stream from a URL and saves it to a file.
Args:
url: The URL of the video stream.
filename: The name of the file to save the video to.
"""
# Set chunk size for efficient downloading
chunk_size = 1024
try:
response = requests.get(url, stream=True)
response.raise_for_status() # Raise an exception for failed requests
# Check for content type (video or not)
if not response.headers['Content-Type'].startswith('video/'):
print(f"Warning: Content type '{response.headers['Content-Type']}' might not be a video.")
with open(filename, 'wb') as f:
for chunk in response.iter_content(chunk_size):
# Filter out keep-alive new chunks
if chunk:
f.write(chunk)
print(f"Download complete! Saved to: {filename}")
except requests.exceptions.RequestException as e:
print(f"Error downloading video: {e}")
# Function to download a video from YouTube using pytube
def download_video_by_pytube(url, base_path="//content//drive//MyDrive//YouTube//Download_video//",
playlist_start=1,
dl_pytube_stream=True,
skip_download=False,
max_resolution=1280,
target_languages=["en"], translated_lang=[]):
# Check if the URL is for a playlist
if 'list=' in url.lower():
playlist = Playlist(url)
folder_name = f"{playlist.title}_{playlist.owner}"
playlist = list(playlist.video_urls)
else:
playlist = []
playlist.append(url)
youtube = YouTube(url)
folder_name = f"{youtube.title}_{youtube.author}"
for i, video_url in enumerate(playlist[playlist_start-1:]):
youtube = YouTube(video_url)
# Get the name of the video
title = youtube.title.translate(table_punctuation).translate(table_digit)
j = str(i+playlist_start).zfill(3)
filename = f"{j}_{title}"
print(f"{int(j)}-th from {len(playlist)} : {title}")
filename=f"{filename}.mp4"
try:
if not dl_pytube_stream:
video_dict = {}
for streaming_data in youtube.streaming_data['formats']:
if 'video/mp4;' in streaming_data['mimeType']:
video_dict[streaming_data['url']] = int(streaming_data['qualityLabel'].split('p')[0])
video_dict = filtered_dict = dict(filter(lambda x: x[1] <= max_resolution, video_dict.items()))
videos = sorted(video_dict.items(), key=lambda x: x[1], reverse=True)
# print(videos[0][0])
download_stream(videos[0][0], filename)
else:
# Get the streams that have a resolution
streams = [stream for stream in youtube.streams.filter(progressive=True, file_extension='mp4') if stream.resolution]
print('************************')
# Sort the streams by resolution in descending order
streams.sort(key=lambda stream: int(stream.resolution.replace('p', '')), reverse=True)
# Find the first stream that has a resolution less than or equal to max_resolution
stream = next((stream for stream in streams if int(stream.resolution.replace('p', '')) <= max_resolution), None)
# Download the video and subtitle and specify the filename
if stream:
if not skip_download:
stream.download(filename)
print("Downloaded resolution is:", stream.resolution)
print('_'*50)
except Exception as e:
print(e)you must set |
|
+1 |
|
No problema, @mahdjourOussama ⭐ |
thanks alot! finally got it working |
|
please merge the fix |
|
+1 can confirm bug |
|
+1 I'm getting the same bug in collab as well. |
|
+1 I'm getting the same bug in collab |
|
If you just want to monkey patch in the fix, put this file somewhere and then import it before you use pytube. It's a copy of the normal pytube Most of the file isn't necessary, and is just there because it's a straight copy of |
working perfectly afeter session reboot in collab |
|
For those who are trying to solve this in their Docker Container (I suppose it works for Google Collab as well but I'm not sure):
|
Updated the following code: Comment out line # 30: var_regex = re.compile(r"^\w+\W") Added following line: var_regex = re.compile(r"^[\w$_]+\W") This fixes the following issue: pytube#1918
RegexMatchError: __init__: could not find match for ^\w+\W - see pytube#1918
It worked perfectly! |
Not for me though somehow, so i made this code from pytube import cipher
def fix(self, js: str):
import re #this is added since re isn't imported
self.transform_plan: List[str] = cipher.get_transform_plan(js) #added "cipher." to every function below to allow it to be called
var_regex = re.compile(r"^[\w\$_]+\W") #the line 30 modification is done
var_match = var_regex.search(self.transform_plan[0])
if not var_match:
raise RegexMatchError(
caller="__init__", pattern=var_regex.pattern
)
var = var_match.group(0)[:-1]
self.transform_map = cipher.get_transform_map(js, var)
self.js_func_patterns = [
r"\w+\.(\w+)\(\w,(\d+)\)",
r"\w+\[(\"\w+\")\]\(\w,(\d+)\)"
]
self.throttling_plan = cipher.get_throttling_plan(js)
self.throttling_array = cipher.get_throttling_function_array(js)
self.calculated_n = None
cipher.Cipher.__init__=fixand this worked for me. Remember to import it in a global environment and dont import cipher (or from pytube import *) again |
Worked for me, too. Running on Kaggle. Thanks! |
thanks, this method works fine and seems more stable |
I created new file with name custom_cipher & change the class Cipher to CustomCipher after that I initialise the cipher.Cipher = CustomCipher but in docker it is not getting call to my CustomCipher class. Please suggest. |
|
سلام
…On Wed, May 8, 2024 at 12:27 AM Renata Leite ***@***.***> wrote:
+1 I'm getting the same bug in collab
—
Reply to this email directly, view it on GitHub
<#1918 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AC5FUQQ532IVBWWAWJ5LEJDZBE54TAVCNFSM6AAAAABHJUWU26VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDAOJZGI4TQNZXGI>
.
You are receiving this because you commented.Message ID:
***@***.***>
--
***@***.***
|
Add this at the top of your code and it will override the cipher. I made it because i dont want to modify any codes in pytube package. Make sure the cipher is imported globally (i.e. not inside a sub-program or class but in the top of your code) and it will work. from pytube import cipher
def fix(self, js: str): #this is __init__(self, js: str) copied from the class Cipher
import re #this line is added since re isn't imported
self.transform_plan: List[str] = cipher.get_transform_plan(js) #added "cipher." to every function below to allow it to be correctly called
var_regex = re.compile(r"^[\w\$_]+\W") #the line 30 modification is done here
var_match = var_regex.search(self.transform_plan[0])
if not var_match:
raise RegexMatchError(
caller="__init__", pattern=var_regex.pattern
)
var = var_match.group(0)[:-1]
self.transform_map = cipher.get_transform_map(js, var) #added "cipher."
self.js_func_patterns = [
r"\w+\.(\w+)\(\w,(\d+)\)",
r"\w+\[(\"\w+\")\]\(\w,(\d+)\)"
]
self.throttling_plan = cipher.get_throttling_plan(js) #added "cipher."
self.throttling_array = cipher.get_throttling_function_array(js) #added "cipher."
self.calculated_n = None
cipher.Cipher.__init__=fix #override class Cipher's __init__ with my fix
#Your own code starts here... |
if you face this issue i think it better that you fork the rep and change line 33 in the cipher file and then instead when downloading from your repo like this |
|
A new release is the best solution. |
and others
I have changed the cipher.py as suggested by many past issues but it doesnt fix it for me.
The text was updated successfully, but these errors were encountered: