How to determine the article PDF download link from the article webpage?

Question

I would like to download some articles from my DOI list automatically (about 1500). Using doi.org I can get a website content for every of them. But the problem is that every website is unique and I do not know how to determine a download link amongst a number of hrefs. Please, could you suggest anything useful for such aim in Python?

P. S. The speech is about free access articles. So I can be sure that the link exists.

Answer 1

As it turned out, the most convenient way is to use metapub library. Note, that it demands Visual Studio C++ 2015 and recenter.

import metapub
from urllib.request import urlretrieve

def downloadByDOI(doi, handle):
    def download(url, handle):
        try:
            urlretrieve(url, handle)
        except:
            download(url, handle)
    
    url = metapub.FindIt(doi=doi).url
    download(url, handle)