Add method extract_info to FileDownloader

[jaimeMF]

That method would look like this:

def extract_info(self,url):
    #iterate over info extractors
    #and return the list of videos

Then in info extractors instead of calling fd.download([url]), we could call fd.extract_info(url) and append the videos returned to the ones we've already got.
This makes InfoExtractors just get the videos info. If needed file downloader will download them.
It would make easier to use youtube-dl as an API for getting videos info.

the download method would need to be rewritten:

def download(self,urls):
    for url in urls:
    videos=self.extract_info(url)
    #actually download the videos

[phihag]

Yeah, that would be nice, although at that point, I'd prefer a YoutubeDL class (along with Pythonic module-level methods as a shortcut to it), that setups openers, constructs IEs, handles I/O, and constructs the FileDownloader.

[jaimeMF]

So, you'd prefer a class for getting the videos info and other for downloading?

[phihag]

No, one to coordinate everything, with methods like extract, download, postprocess, to_screen and one to actually download.

[jaimeMF]

Understood.
I think making IE just get only the info is the easiest part, I'll have a look.

[jaimeMF]

Done