def scrape(abstract_url):
tree, urls, page_text = get_tree(abstract_url)
article = make_blank_article()
article['scraper'] = 'rup'
article['source_urls'] = [uri for _, uri in urls]
article['title'] = get_meta('DC.Title', tree)
article['journal'] = get_meta('citation_journal_title', tree)
article['publisher'] = get_meta('DC.Publisher', tree)
article['author_names'] = get_meta_list('DC.Contributor', tree)
article['abstract'] = strip_space( tree.xpath("//div[@class='section abstract']/p")[0].text_content() )
article['citation']['journal'] = get_meta('citation_journal_abbrev', tree)
article['citation']['volume'] = get_meta('citation_volume', tree)
article['citation']['page'] = get_meta('citation_firstpage', tree)
article['ids'] = {'doi':get_meta('DC.Identifier', tree),}
pub_date = get_meta('DC.Date', tree)
if pub_date:
split = pub_date.split('-')
article['date_published'] = make_datestamp(split[2], split[1], split[0])
article['citation']['year'] = split[0]
return article
def scrape(abstract_url):
tree, urls, page_text = get_tree(abstract_url)
article = make_blank_article()
article['scraper'] = 'rsc'
article['source_urls'] = [uri for _, uri in urls]
article['ids'] = {'doi': get_meta('DC.Identifier', tree), }
article['title'] = get_meta('DC.title', tree)
article['publisher'] = get_meta('DC.publisher', tree)
article['author_names'] = get_meta_list('DC.Creator', tree)
try:
article['journal'] = tree.xpath("//img[@id='imgLoader']/@title")[0]
except:
pass
try:
article['abstract'] = tree.xpath("//p[@xmlns='http://www.rsc.org/schema/rscart38']")[0].text_content()
except:
pass
article['citation']['journal'] = get_meta('citation_journal_title', tree)
article['citation']['volume'] = get_meta('citation_volume', tree)
article['citation']['page'] = get_meta('citation_firstpage', tree)
pub_date = get_meta('citation_publication_date', tree)
if pub_date:
split = pub_date.split('-')
article['date_published'] = make_datestamp(split[2], split[1], split[0])
article['citation']['year'] = split[0]
return article
def scrape(abstract_url):
tree, urls, page_text = get_tree(abstract_url)
article = make_blank_article()
article['scraper'] = 'MIT'
article['source_urls'] = [uri for _, uri in urls]
article['title'] = get_meta('dc.Title', tree)
article['publisher'] = get_meta('dc.Publisher', tree).strip()
article['author_names'] = get_meta_list('dc.Creator', tree)
# Two identifier schemes used --- do we want both?
article['ids'] = dict(zip(['publisher-id','doi'], get_meta_list('dc.Identifier', tree)))
article['journal'] = tree.xpath("//h1[@class='journalTitle']/a/img/@alt")[0]
# dc.Description is present, but contains an abbreviated abstract
# --- this gets the full abstract
article['abstract'] = tree.xpath("//div[@class='abstractSection']/p/text()")[0]
article['citation']['journal'] = article['journal']
# Citation details (volume, number, pages) given as text
# immediately following the h1 tag.
# example: December 2012, Vol. 24, No. 2, Pages 1-35
citation_text = tree.xpath("//h1[@class='journalTitle']/following-sibling::text()")[0]
pub_year, volume, number, page_first, page_last = re.findall('\d+', citation_text)
article['citation'] = {}
article['citation']['volume'] = volume
article['citation']['number'] = number
article['citation']['page_first'] = page_first
article['citation']['page_last'] = page_last
date = get_meta('dc.Date', tree).split('-')
if date:
article['date_published'] = make_datestamp(date[2], date[1], date[0])
article['citation']['year'] = date[0]
return article
def scrape(abstract_url):
req = urllib2.Request(abstract_url, headers=utils.headers)
urls, response = utils.get_response_chain(req)
if response.info().get('Content-Encoding') == 'gzip':
buf = StringIO( response.read())
f = gzip.GzipFile(fileobj=buf)
data = f.read()
else:
data = response.read()
page_text = data.decode('utf-8')
tree = lxml.html.fromstring(page_text)
article = make_blank_article()
article['scraper'] = 'npg'
article['source_urls'] = [uri for _, uri in urls]
article['title'] = get_meta('DC.title',tree)
if article['title'] == None:
article['title'] = get_meta('dc.title',tree)
article['publisher'] = get_meta('DC.publisher',tree)
if article['publisher'] == None:
article['publisher'] = get_meta('dc.publisher',tree)
article['author_names'] = get_meta_list('DC.creator',tree)
if article['author_names'] == None:
article['author_names'] = get_meta_list('dc.creator',tree)
article['abstract'] = get_meta('description', tree)
if not article['abstract']:
try:
article['abstract'] = tree.xpath("//div[@class='content']/p")[0].text_content()
except:
pass
if not article['abstract']:
try:
article['abstract'] = tree.xpath("//div[@id='abs']/p")[0].text_content()
except:
pass
article['citation']['journal'] = get_meta('citation_journal_title', tree)
article['citation']['volume'] = get_meta('prism.volume', tree)
article['citation']['page'] = get_meta('prism.startingPage', tree)
article['journal'] = get_meta('prism.publicationName', tree)
year = get_meta('citation_date', tree)
if year:
article['citation']['year'] = year[0:4]
article['ids'] = dict(zip(['doi'], [tree.xpath("//meta[@name='citation_doi']/@content")[0][4:]]))
pub_date = get_meta('DC.date', tree)
if pub_date == None:
pub_date = get_meta('dc.date', tree)
if pub_date:
split = pub_date.split('-')
article['date_published'] = make_datestamp(split[2], split[1], split[0])
#Specific abstract scrapers for subsidiary journals
if article['journal'] == 'The EMBO Journal':
try:
article['abstract'] = tree.xpath("//p[@class='lead']")[0].text_content()
except:
pass
elif article['journal'] == 'EMBO reports':
try:
article['abstract'] = tree.xpath("//p[@class='lead']")[0].text_content()
except:
pass
elif article['journal'] == 'Oncogene':
try:
article['abstract'] = tree.xpath("//p[@class='abs lead']")[0].text_content()
except:
pass
return article
def scrape_tree(tree, urls, page_text):
article = {}
article['source_urls'] = [uri for _, uri in urls]
article['scraper'] = 'default'
article['title'] = get_meta(title, tree)
article['author_names'] = get_meta_list(authors, tree)
if article['author_names'] is None:
article['author_names'] = get_meta(authors_list, tree).split(';')
article['abstract'] = get_meta(abstract, tree)
article_doi = get_meta(doi, tree)
if article_doi:
article_doi = article_doi.replace('http://dx.doi.org/','')
article['ids'] = {'doi': article_doi}
article['journal'] = get_meta(journal, tree)
article['publisher'] = get_meta(publisher, tree)
article['citation'] = {}
article['citation']['journal'] = article['journal']
article['citation']['volume'] = get_meta(volume, tree)
article['citation']['page_first'] = get_meta(firstpage, tree)
article['citation']['page_last'] = get_meta(lastpage, tree)
article['keywords'] = get_meta(keywords, tree)
date = get_meta(publication_date, tree)
if date:
date = parse_date(date)
article['date_published'] = time.mktime(date.timetuple())
article['citation']['year'] = date.year
return article