[Bug-wget] Re: Fwd: Downloading video list from youtube profile.

[augustin]

On Friday 31 October 2008 19:41:20 you wrote:
> [Micah forwarded your mail to the wget list.]
>
> I'd suggest using clive (http://clive.sf.net) rather than wget when
> dealing with youtube, google video, et al.
>
> :; clive --scan
> : 'http://tw.youtube.com/profile?user=BarackObamadotcom&view=videos'
>
> gives a list of the available videos and lets one download them.
>
> Grabbing a single video is also easy:
> :; clive http://youtube.com/watch?v=Mg56KbtmARc
>
> clive 1.0.2 20081014  [Linux]
> http://youtube.com/watch?v=Mg56KbtmARc&fmt=18                          
> (cache) RobotsAttack.mp4                                                   
>       3.3MB => 1 (3.3MB), failed: 0, skipped: 0
> RobotsAttack.mp4                              100%    3.3MB  350.0KB/s
> 00:00:09
>
> Clive is likely available in your distribution if linux or bsd, else the
> website has download links to the code.


Hello James,

Thank you for your reply.

I already had clive installed, but only version 0.5 which does not have 
the --scan option.
I have downloaded the latest stable version, and battled a couple of hours to 
get all the perl dependencies, after which I have been able to try 
clive --scan.

Clive looks cool. I'm sure I'm going to use it, now. Thanks.

I had been using this to download specific videos:
http://www.arrakis.es/~rggi3/youtube-dl/
It's simple and I was happy with it.

However, what I need is a list of all videos in a channel. As I tried to 
explained, I'm looking for a way to programmatically browser ALL the pages of 
videos in a channel.

I just tried clive as you suggested, but it only gives the 20 latest videos in 
the channel. However, the Obama channel for example has close to 1000 pages 
of videos (20 videos per page). How do I get the LINKS to the videos listed 
on page 2 onwards? 

I was trying to reproduce the AJAX call with wget, so that I can get the HTML 
with the list of videos on the 2nd, 3rd page, etc.
I could then parse the HTML and use clive or youtube-dl to get the video 
proper.

As I tried to explain in my first email, I have all the specifics of the HTTP 
request to the youtube server thanks to firefox + the firebug extension. I'm 
still testing how to reproduce that with wget.

Meanwhile, it could be worthwhile to contact the clive author to add the same 
feature in his program. 

I'd still would like to know the solution with wget, because it's more 
versatile, and if I know how to do it with wget, I can reproduce something 
similar to get information from other sites.


Thanks for your help.


Augustin.

http://overshoot.tv/