def add_missing(query, max_count, randomize=False):
"""Search PubMed for articles and scrape documents.
:param str query: PubMed query
:param int max_count: Maximum number of articles to process
:param bool randomize: Randomize list of articles to fetch
:return: Added article objects
"""
pmids = pubtools.search_pmids(query)
stored_pmids = [
article['pmid'] for article in mongo['article'].find({}, {'pmid': 1})
]
missing_pmids = set(pmids) - set(stored_pmids)
logger.warn('Found {0} articles to add.'.format(len(missing_pmids)))
pmids_to_add = list(missing_pmids)[:max_count]
if randomize:
random.shuffle(pmids_to_add)
records = pubtools.download_pmids(pmids_to_add)
scraper = SCRAPE_CLASS(**SCRAPE_KWARGS)
added = []
for pmid, record in zip(pmids_to_add, records):
logger.debug('Adding article {}'.format(pmid))
article = Article.from_record(record)
article.scrape(scraper)
article.tag()
added.append(article)
return added
def update(cls, query, max_count):
"""Search PubMed for articles and scrape documents.
:param str query: PubMed query
:param int max_count: Maximum number of articles to process
:return list: Added article objects
"""
pmids = pubtools.search_pmids(query)
stored_pmids = [
article['pmid']
for article in mongo['article'].find(
{}, {'pmid': 1}
)
]
pmids_to_add = set(pmids) - set(stored_pmids)
pmids_to_add = list(pmids_to_add)[:max_count]
records = pubtools.download_pmids(pmids_to_add)
scraper = SCRAPE_CLASS(**SCRAPE_KWARGS)
added = []
for pmid, record in zip(pmids_to_add, records):
logging.debug('Adding article {}'.format(pmid))
article = Article.from_record(record)
article.scrape(scraper)
added.append(article)
return added
def test_download_pmids(self):
""" Downloading articles by PMID should return correct MEDLINE data. """
records = pubtools.download_pmids(['1', '2', '3'], chunk_size=2)
assert_equal(len(records), 3)
assert_equal(records[0], DATA_PMID[0])
assert_equal(records[1], DATA_PMID[1])
assert_equal(records[2], DATA_PMID[2])
def test_download_pmids(self):
""" Downloading articles by PMID should return correct MEDLINE data. """
records = pubtools.download_pmids(["1", "2", "3"], chunk_size=2)
assert_equal(len(records), 3)
assert_equal(records[0], DATA_PMID_1)
assert_equal(records[1], DATA_PMID_2)
assert_equal(records[2], DATA_PMID_3)
def _resolve_pmid(self, pmid):
"""Follow PMID link, store HTML, and return final URL."""
# Get DOI from PubMed API
pub_data = pubtools.download_pmids([pmid])[0]
doi = pubtools.record_to_doi(pub_data)
if doi:
return self._resolve_doi(doi)
pub_link = pubtools.pmid_to_publisher_link(pmid)
# Follow publisher link
if pub_link:
# Browse to link
self.browser.open(pub_link)
# Read documents and save in ScrapeInfo
self.info.init_html, self.info.init_qhtml = self.browser.get_docs()
# Return URL
return self.browser.geturl()
def add_missing(query, max_count, randomize=False):
"""Search PubMed for articles and scrape documents.
:param str query: PubMed query
:param int max_count: Maximum number of articles to process
:param bool randomize: Randomize list of articles to fetch
:return: Added article objects
"""
pmids = pubtools.search_pmids(query)
stored_pmids = [
article['pmid']
for article in mongo['article'].find(
{}, {'pmid': 1}
)
]
missing_pmids = set(pmids) - set(stored_pmids)
logger.warn('Found {0} articles to add.'.format(len(missing_pmids)))
pmids_to_add = list(missing_pmids)[:max_count]
if randomize:
random.shuffle(pmids_to_add)
records = pubtools.download_pmids(pmids_to_add)
scraper = SCRAPE_CLASS(**SCRAPE_KWARGS)
added = []
for pmid, record in zip(pmids_to_add, records):
logger.debug('Adding article {}'.format(pmid))
article = Article.from_record(record)
article.scrape(scraper)
article.tag()
added.append(article)
return added
def from_pmid(cls, pmid):
records = pubtools.download_pmids([pmid])
return cls.from_record(records[0])
def from_pmid(cls, pmid):
records = pubtools.download_pmids([pmid])
return cls.from_record(records[0])
def test_download_pmids(self):
""" Downloading articles by PMID should return correct MEDLINE data. """
with assert_raises(pubtools.EntrezEmailError):
pubtools.download_pmids(['1', '2', '3'], chunk_size=2)
