def save(self, name, save_dir='.', save_dirs=None):
"""
"""
save_dirs = save_dirs or {}
saved = {}
# Write documents
for doc_type in self.docs:
doc_save_dir = save_dirs.get(doc_type, save_dir)
# Make directory if necessary
if doc_save_dir != '.':
utils.mkdir_p(doc_save_dir)
file_name = '{}.{}'.format(name, EXTENSIONS[doc_type])
save_name = os.path.join(doc_save_dir, file_name)
with open(save_name, 'w') as f:
f.write(self.docs[doc_type])
saved[doc_type] = {
'name': file_name,
'path': doc_save_dir,
}
return saved
def searchscrape(query, out_dir, scrape_klass=scrape.Scrape, **kwargs):
'''Run PubMed query, then download articles.
Args:
query (str) : PubMed search query
out_dir (str) : Output directory for documents
scrape_klass (Scrape) : Scraper type
kwargs: Optional arguments for PubMed search
Returns:
None
Examples:
>>> searchscrape('fmri AND neuroimage[journal]', '.', retmax=5)
'''
# Create directory if needed
utils.mkdir_p(out_dir)
# Find articles on PubMed
searcher = pubtools.PubMedSearcher()
pmids = searcher.search(query, **kwargs)
# Initialize scraper
scraper = scrape_klass()
# Loop over articles
for pmid in pmids:
print 'Working on article %s...' % (pmid)
# Scrape article documents
info = scraper.scrape(pmid=pmid)
# Write documents to files
for doc_type in info.docs:
out_name = '%s/%s.%s' % (out_dir, pmid, doc_type)
with open(out_name, 'w') as f:
f.write(info.docs[doc_type])
