class DOI(DOIMetadata):
"""find and download a pdf for the doi given
>>> DOI('10.1016/0166-218X(92)00170-Q')
"""
headers = generic_headers
headers['Host'] = 'gen.lib.rus.ec'
headers['Referer'] = 'http://gen.lib.rus.ec/scimag/'
url = 'http://gen.lib.rus.ec/scimag/?s={}&journalid=&v=&i=&p=&redirect=1'
def __init__(self, doi, *args, **kwargs):
super().__init__(doi, *args, **kwargs)
self.url = URIRef(self.url.format(self.doi))
r = requests.get(self.url, headers=self.headers)
r.raise_for_status()
self.links = re.compile(r'a href="([^"]+pdf)"').findall(r.text)
link, *links = self.links
r = requests.get(link, stream=True)
self.filename = Literal('.'.join((doi.replace('/','_'), 'pdf')))
with open(self.filename, 'wb') as fd:
for chunk in r.iter_content(1024*10):
fd.write(chunk)
self.path = URIRef(urljoin('file:', pjoin(os.getcwd(), self.filename)))
self.graph.add((self.path, URIRef('http://purl.org/dc/terms/identifier'), self.doi))
self.graph.commit()
