Anyone scrape Youtube with curl

[Konroy pp]

Hello,
I dont want use the API because of the restrictions. Is someone scrap Youtube for video with PHP and curl? I have some ipv6/4 proxy, is that enough?
Thanks!

 

[meannn]

Was doing 8 years ago, you are late.... try using with socks 5 proxies.

 

[Konroy pp]

Thanks man, donc you have any ressources ?

 

[shreyamalhotra]

you can use ubot to scrape fast

 

[Gogol]

Short answer, possible.

Long answer, it is much tougher than just downloading with curl. Think about getting the actual video URL first (Just get the first chunk.) . Then analize the URL and modify start byte, end byte, duration etc and then use that modified URL to be downloaded with curl. Also, because you are dealing with proxies; make sure that you download from the same IP you used to get the video download URL. AFAIK, those URLs are IP and time sensitive.

 

[Konroy pp]

Thanks for your response, actually I dont want download because I already have a script, I just want script which scrape search result

 

[xoskillet]

You could just go download AutoIt, then do a google search for scraping youtube or google with autoit. You'll find working examples. Copy/Run that code... Do whatever little mods you need to do, and you'll have your scraper built in less than 30 minutes.

 

[TasDePixels]

You can easly bypass rate limit with multiple API keys.
I personaly use 5 api keys for real time monitoring, no need to for scraping.

 

[ChillerNiller]

Couple of months ago I was also looking for a scraper in PHP. Turns out, nowadays you can't do it - too much javascript.

Found this gem: https://github.com/GoogleChrome/puppeteer
Its written in NodeJS, but you can easily make a wrapper/API around it and call it from PHP.

 

[Solixy]

As Tas said, just get multiple keys. The process through the API is going to be much easier than scraping. Also, don't be afraid to apply for a rate increase. I applied for one and got approved for 200,000 queries a day.

 

[ahriyu]

As Tas said, just get multiple keys. The process through the API is going to be much easier than scraping. Also, don't be afraid to apply for a rate increase. I applied for one and got approved for 200,000 queries a day.

How did you do ? didn't they ask you for "fully-functional demo accounts, screenshots and documents in relation to your API Client" before increasing ?

 

[Solixy]

How did you do ? didn't they ask you for "fully-functional demo accounts, screenshots and documents in relation to your API Client" before increasing ?

I actually do upload videos from the API on a daily basis for automated youtube channel using youtubeuploader by Porjo on Github (I can't post the link since I have < 15 posts). All I did was make a video uploading a video using his API client and linked his GitHub page when they asked for supporting documentation and screenshots and it was approved .